Romer V. Evans: a Legal and Political Analysis Caren G

Law & Inequality: A Journal of Theory and Practice

Volume 15 | Issue 2 Article 1

1997 Romer v. Evans: A Legal and Political Analysis Caren G. Dubnoff

Follow this and additional works at: http://scholarship.law.umn.edu/lawineq

Recommended Citation Caren G. Dubnoff, Romer v. Evans: A Legal and Political Analysis, 15 Law & Ineq. 275 (1997). Available at: http://scholarship.law.umn.edu/lawineq/vol15/iss2/1

Law & Inequality: A Journal of Theory and Practice is published by the University of Minnesota Libraries Publishing. Romer v. Evans: A Legal and Political Analysis

Caren G. Dubnoff*

Introduction Despite the Supreme Court's role as final arbiter of the "law of the land," its power to effect social change is limited. For example, school desegregation, mandated by the Court in 1954, was not actually implemented until years later when Congress and the President finally took action.1 As a result, prayer in public schools, repeatedly deemed illegal by the Court, continues in many parts of the country even today. 2 To some degree, whether the Court's po-

* Associate Professor, Department of Political Science, College of the Holy Cross. Ph.D. 1974, Columbia University; A.B. 1964, Bryn Mawr. The author wishes to thank Jill Moeller for her most helpful editorial assistance. 1. Several studies have demonstrated that Brown v. Board of Education, 347 U.S. 483 (1954), produced little school desegregation by itself. One of the earliest of these was J.W. PELTASON, FIFTY-EIGHT LONELY MEN: SOUTHERN FEDERAL JUDGES AND SCHOOL DESEGREGATION (1961) (demonstrating how district court judges evaded the decision, leaving school segregation largely in place). For a more recent analysis, see GERALD N. ROSENBERG, THE HOLLOW HOPE 46-49 (1991) (observing that Brown could not be enforced until Congress and the President joined the effort in 1964 with the passage of the 1964 Civil Rights Act). But see CHARLES A. JOHNSON & BRADLEY C. CANON, JUDICIAL POLICIES: IMPLEMENTATION AND IMPACT 257 (1984) (arguing that though Brown's direct effect was necessarily limited because the Court lacks enforcement powers, the decision had a significant psychological impact). 2. The Court's initial pronouncements on this issue came in Engel v. Vitale, 370 U.S. 421 (1962) (striking down a prayer written by New York officials for reci- tation in the New York public schools), and Abington School District v. Schempp, 374 U.S. 203 (1963) (invalidating state-mandated Bible reading in Pennsylvania public schools). Following these decisions, compliance with the Court's prohibition of religious activities in public schools varied across the country. See Robert H. Birkby, The Supreme Court and the Bible Belt: Tennessee Reaction to the 'Schempp" Decision, in 1 PRAYER IN THE PUBLIC SCHOOLS AND THE CONSTITUTION 1961-1992 440 (Robert Sikorski ed., 1993) (analyzing possible reasons for Tennes- see schools' partial compliance with Schempp); KENNETH M. DOLBEARE & PHILLIP E. HAMMOND, THE SCHOOL PRAYER DECISIONS: FROM COURT POLICY TO LOCAL PRACTICE (1971) (investigating the failure of five towns in one midwestern state to conform to the Court's outlawing of prayer in public schools); H. Frank Way, Jr., Survey Research on Judicial Decisions: The Prayer and Bible Reading Cases, in 1 PRAYER IN THE PUBLIC SCHOOLS AND THE CONSTITUTION 1961-1992, supra, at 457 (finding proscribed religious practices in public schools, with the exception of Southern schools, had largely disappeared). But see WILLIAM K. MUIR, JR., Law and Inequality [Vol. 15:275 sition prevails in the long run is beyond its control. Presidents may respond to the Court's actions in ways that are helpful or in ways that impede realization of the Court's objectives. 3 Moreover,

PRAYER IN THE PUBLIC SCHOOLS: LAW AND ATTITUDE CHANGE 113-17 (1967) (showing substantial implementation of the Court's mandate to remove prayer from public schools when the school board was prodded by a local lawyer). For further discussions regarding uneven judicial compliance with the Su- preme Court's prohibition on religious activities in schools, see JOHNSON & CANON, supra note 1, at 65-66 (discussing possible reasons for noncompliance by lower federal judges); KENNETH D. WALD, RELIGION AND POLITICS IN THE UNITED STATES 149-50 (3d ed. 1997) (discussing how the religious backgrounds of judges affect their decisions regarding separation of church and state). Express noncompliance continues to be a problem, as is demonstrated in the case of one Kentucky principal who led his students in Christian prayer over the public announcement system. See Attacks on the Freedom to Learn, PRESS RELEASE (People for the American Way, Washington, D.C.), Sept. 4, 1996. In another instance, the American Civil Liberties Union (ACLU), acting on behalf of parents and students in Santa Fe, Texas, filed suit against religious exercises in the public schools including grade school teacher-led prayers. ACLU of Texas Files a Complaint Against Santa Fe ISD, PRESS RELEASE (ACLU of Texas), Apr. 4, 1995. Other school districts have attempted to avoid the Court's decisions, which following Lee v. Weisman, 505 U.S. 577 (1992), include a prohibition on school- initiated graduation prayers by encouraging student initiated prayer. So far, policies permitting students to lead prayer during regular school hours have been struck down by the courts. See, e.g., Ingebretsen v. Jackson Pub. Sch. Dist., 88 F.3d 274 (5th Cir. 1996) (invalidating a Mississippi statute that permitted public school students to initiate nonsectarian, nonproselytizing prayer at compulsory and noncompulsory school events); Herdahl v. Pontotoc County Sch. Dist., 887 F. Supp. 902 (N.D. Miss. 1995) (enjoining a school district's practice of allowing student organizations to broadcast prayers over the school intercom and to initiate student-led prayer during school hours). 3. The President's ability to alter judicial policy through appointments de- pends on how many appointments can be made, the divisions on the Court, attention given to ideology in making appointments and the willingness of the Senate to allow the President leeway in appointments. See JOHNSON & CANON, supra note 1, at 161-62. Presidents at times have turned the Court's decisions into a political issue. In 1968, Richard Nixon campaigned against the Court's criminal justice decisions. See id. at 3-4. Nixon then used his appointment power to place men on the Court who agreed with his position. See generally LEONARD W. LEVY, AGAINST THE LAW: THE NIXON COURT AND CRIMINAL JUSTICE (1974) (detailing the profound changes in criminal procedure brought about by Nixon appointees). Presidents can affect judicial policies in other ways as well. Failure to endorse the Court's position may provide encouragement to those who would resist it, as Eisenhower's silence following Brown v. Board of Education apparently stalled the process of desegregation. See PELTASON, supra note 1, at 46-55. For example, Eis- enhower may have inadvertently encouraged Governor Faubus' resistance to desegregation in Little Rock. JOHNSON & CANON, supra note 1, at 157-58. Clinton may have helped defuse the political issue generated in the Romer decision by agreeing to the initiative barring same sex marriage, though it is believed that Clinton's motive for supporting the ban on same sex marriages was to "blunt[ ] an issue the Republicans had hoped to use against him in the election," not to influence the Court. The Senate's Vote on Gay Rights, N.Y. TIMES, Sept. 13, 1996, at A34. 1997] ROMER V. EVANS presidential appointments can affect the composition of the Court 4 and allow earlier decisions to be modified or reversed. Even though the Court does not have the final word on enforcement, it can act in ways that maximize its influence and strengthen its power. The Court's power is weakest when its decision is opposed to a broad national consensus on a matter of policy.5 Its influence is also diminished when its decisions are weakly argued or ambiguous because this leaves the Court vulnerable to professional criticism which might later be exploited by political opponents. 6 Finally, for cases which have a strong political dimen- sion,7 the Court weakens its position when it fails to address those objections to its action most likely to resonate in the public de- 8 bate.

4. See generally LAURENCE TRIBE, GOD SAVE THIS HONORABLE COURT: How THE CHOICE OF SUPREME COURT SHAPES OUR HISTORY (1985) (examining the importance of presidential appointments on the direction of the law). Despite the importance of presidential backing, the President is but one political actor. The school prayer experience demonstrates the significance of support from other participants in the political process. In the aftermath of Brown and its progeny, for example, it mattered a great deal whether local political leaders spoke out for compliance or instead looked the other way as school authorities followed their own preferences. See William M. Beaney & Edward N. Beiser, Prayerand Politics: The Impact of Engel and Schempp on the Political Process, in 1 PRAYER IN THE PUBLIC SCHOOLS AND THE CONSTITUTION 1961-1992, supra note 2, at 411. 5. See DAVID M. O'BRIEN, 1 CONSTITUTIONAL LAW AND POLITICS, STRUGGLES FOR POWER AND GOVERNMENT ACCOUNTABILITY 180 (2d ed. 1995) (arguing that public opinion constrains the Court from going too far or too fast). The judgment that the Court hurts itself by going "too far, too fast" is apparently shared by Jus- tice Ruth Bader Ginsburg. See Linda Greenhouse, An Issue for a Reluctant High Court, N.Y. TIMES, Oct. 6, 1996, at E3; see also Thomas R. Marshall, The Rehnquist Court and Public Opinion: The First Decade, Paper Given at the Ameri- can Political Science Association Annual Meeting (Aug. 31-Sept. 2, 1996) (on file with Law & Inequality: A Journalof Theory and Practice)(offering evidence that a slight majority of the Rehnquist Court's rulings reflect majority popular opinion); William Mishler & Reginald S. Sheehan, The Supreme Court as a Countermajori- tarian Institution?: The Impact of Public Opinion on Supreme Court Decisions, 87 AM. POL. SCI. REV. 87 (1993) (arguing that there was a strong link between public opinion and the Court's decisions from 1956-1981 that was later weakened by the strongly ideological appointments of Presidents Reagan and Bush). But see Hel- mut Norpoth & Jeffrey Segal, Popular Influence on Supreme Court Decisions, 88 AM. POL. SCI. REV. 711 (1994) (challenging Mishler & Sheehan's methodology and conclusions). 6. See PELTASON, supra note 1, at 18 (attributing the failure to enforce desegregation at least in part to the ambiguity and reasoning of Brown). 7. I refer here to those cases which receive wide attention in the media and become the focus of partisan and interest group politics. Roe v. Wade, 410 U.S. 113 (1973), is the most obvious example. 8. See, e.g., GUIDO CALABRESI, IDEAS, BELIEFS, ATTITUDES, AND THE LAW 93- 114 (1985) (arguing that an approach to Roe v. Wade taking account of the concerns of those who opposed abortion would have generated more acceptance); LIEF H. CARTER, REASON IN LAW 229-33 (4th ed. 1994) (arguing that the judiciary gains 278 Law and Inequality [Vol. 15:275

This Comment analyzes one of the more highly-charged cases of the 1996 term, Romer v. Evans,9 from both a legal and a political perspective. In Romer, the U.S. Supreme Court struck down an amendment to the Colorado Constitution (Amendment 2) prohibiting all legislative, executive and judicial action designed to protect homosexuals.' 0 Amendment 2, which was approved in a statewide referendum, overruled local laws that barred discrimination based on sexual orientation.1 ' Rejecting Colorado's assertion that Amendment 2 merely prevented gays, lesbians or bisexuals from obtaining special privileges under state law, the Court's 6- 3 majority held that the law was discriminatory. 12 The Court reasoned that the Fourteenth Amendment prohibits states from broadly disadvantaging an identifiable group solely because the group is unpopular.'3 The Court held that Amendment 2 violates the Equal Protection Clause because a state may not make a "class of persons a stranger to its laws."'14 In dissent, Justice Scalia (joined by Chief Justice Rehnquist and Justice Thomas) accused the majority of taking sides in "cultural debate[s]" and of exercis- ing power in a way contrary to democratic decision making.15

legitimacy and authority through good legal reasoning which includes an open weighing of the arguments on both sides of an issue). But see JOHNSON & CANON, supra note 1, at 208 (arguing that there is little empirical evidence that supports the proposition that reasoning matters). 9. 116 S. Ct. 1620 (1996). 10. Id. at 1623. The author recognizes that Amendment 2 was intended to eliminate protections to gay man, lesbians and bisexuals; however, for clarity and brevity, the single term "homosexuals" will be used throughout this Comment. The reader should remain aware that the term is meant to encompass all three of the groups targeted by Amendment 2. 11. Id. Denver, Aspen and Boulder passed legislation that protected individuals from discrimination due to their sexual status. Id. 12. Id. at 1624-27. 13. Id. at 1628. 14. Id. at 1629. 15. Id. A few weeks later, with the same Justices as in Romer comprising a 6-3 vote, the Supreme Court vacated a Sixth Circuit Court of Appeals decision upholding a 1993 Cincinnati referendum which, like Colorado's Amendment 2, repealed any existing laws protecting against discrimination based on sexual orientation and barred new legislation on that subject. Equality Found. of Greater Cincinnati, Inc. v. City of Cincinnati, 116 S. Ct. 2519 (1996), vacating 54 F.3d 261 (6th Cir. 1995). The case was remanded to the lower court for consideration not inconsistent with Romer. Id. Only the dissent, again charging the majority with unduly interfering with democratic decisionmaking, wrote an opinion. Id. (Scalia, J., dissenting). Reaf- firming its opposition to the Romer decision, Justice Scalia nevertheless pointed out that the circumstances in Cincinnati were distinguishable. Id. A decision invalidating Cincinnati's charter amendment would be more intrusive on the democratic process than invalidating a state constitutional amendment since it would overrule a decision by the lowest electoral body. Id. 1997] ROMER V. EVANS

From a legal standpoint, this Comment argues that while the majority had ample justification for its ruling, it did not defend its position as well as it should have, and as a result, Romer is open to future attack. This Comment also takes issue with those observers who claim Romer effected a sweeping revision of constitutional law' 6 and critics who argue that the Court's opinion was undemocratic.17 This Comment concludes that Romer neither provides homosexuals with a broad array of rights nor endorses homosexuality. The Court also did not elevate sexual orientation to a suspect or semi-suspect classification as a guideline for future Four- teenth Amendment litigation. What the Court did do in Romer was re-enter the debate over whether the Constitution had anything to say about how homosexuals should be treated. In that sense, the Court's holding that the State cannot make homosexuals "strangers to its laws"'8 was very important. Homosexuals surely ended up in a better position politically than before the decision. The decision also raised questions about the continuing vi- tality of Bowers v. Hardwick,19 a case in which the Court refused to invalidate a Georgia statute criminalizing sodomy. From a political standpoint, this Comment argues that it was important for the Court's legitimacy for it to decide Romer narrowly, in a way that did not place it far outside majority public opinion. While the Court partially accomplished this task, it could have made an even narrower ruling had it tracked the opinion of the Colorado Supreme Court more closely. 20 A narrower decision would have been just as firmly grounded in law, yet less inviting of future litigation and less subject to political manipulation. It also would have allowed a stronger defense in democratic theory. To the extent the dissent's effort to portray the Court as undemocratic

16. See, e.g., Jan Crawford Greenburg, Court Strikes Anti-Gay Amendment, CHI. TRIB., May 21, 1996, § 1, at 1. 17. See, e.g., Jan Crawford Greenburg, Gay Rights Ruling Short on Guidance, May 22, 1996, § 1, at 4 (quoting Professor Dan Polsby of Northwestern University School of Law). "It opens up all kind of litigation possibilities. I will say this is among the sorriest performances-as an exercise of judicial craftsmanship-that I have ever seen from the Supreme Court." Id.; The Supreme Court Overreaches, CHI. TRIB., May 21, 1996, § 1 (Editorial), at 16 ("[I]t is a defeat for democratic governance that may ultimately inflame rather than erode opposition to gay rights."). 18. Romer, 116 S. Ct. at 1629. 19. 478 U.S. 186 (1986) (upholding a Georgia law criminalizing sodomy challenged under of the Due Process Clause). 20. See Evans v. Romer, 882 P.2d 1335 (Colo. 1994), affig 854 P.2d 1270 (Colo. 1993), afPd, 116 S. Ct. 1620 (1996). For a discussion of the Colorado Supreme Court's reasoning, see infra Part II.B. 280 Law and Inequality [Vol. 15:275 and elitist can be made to stick, there will be negative future effects for the rights of homosexuals and for other minority groups. Part I of this Comment provides a brief discussion of the political setting surrounding Romer. Its purpose is to identify the issues of greatest political salience so that the Court's response to them can be evaluated. Part II sets out the legal issues involved in the decision and summarizes the reasoning of both the majority opinion and of the dissent. Part III provides a legal and political critique of Romer. Finally, Part IV offers suggestions on how the Court could have made a stronger argument, again from the per- spectives of both law and politics. This Comment concludes that although Romer was correctly decided, the majority opinion was not argued as persuasively as it might have been, and this short- coming may have negative future consequences for the civil rights of minorities.

I. Political Setting Romer was about more than the civil rights of homosexuals. It was part of a larger argument over the proper role of religion and family values in public policy making. As with abortion, the 22 gay civil rights debate2' often has led to partisan political feuds. In recent years, a growing number of states and local com- munities have adopted policies prohibiting discrimination based on sexual orientation in public accommodations, employment, insurance and education. 23 The referendum movements, such as the

21. As with "homosexual," the term "gay civil rights" is meant to refer to the legal interests of those targeted by Amendment 2, namely gay men, lesbians and bisexuals. 22. See e.g., Katia Hetter, The New Civil Rights Battle: The Supreme Court Hands Gays a Win in the Struggle Between Tolerance and Tradition,U.S. NEWS & WORLD REP., June 3, 1996, at 28 (anticipating that social conservatives would use the Court's ruling to "energize their grass-roots activists and make homosexuality a wedge issue in the [fall 1996 election] campaign"). Id. Some Republicans saw political gain whichever way the Court went. At the May 3-4 meeting of the New England Political Science Association, Hadley Arkes, an Amherst professor and prominent supporter of the anti-gay initiatives, anticipated that the Court would strike down Amendment 2. He expressed the hope that the decision would be sweeping, in which case he believed that it would pro- duce a realignment toward the Republican party similar to the one provoked by Roe v. Wade, 410 U.S. 113 (1973). 23. Laws barring employment discrimination against homosexuals have been enacted in nine states: California, Connecticut, Hawaii, Massachusetts, Minne- sota, New Jersey, Vermont, Wisconsin and Rhode Island. See Eric Schmitt, Anti- DiscriminationProposal Delays Senate Vote on Bill Opposing Same-Sex Marriage, N.Y. TIMES, Sept. 6, 1996, at A24. A similar measure narrowly failed to pass the Senate. See The Senate's Vote on Gay Rights, supra note 3, at A34. For a thor- ough review of the law in this area, see Robert B. Mison et al., Sexual Orientation 1997] ROMER V. EVANS

one that produced Amendment 2, are strategic attempts by anti- gay forces to reverse this trend.24 In Colorado, the referendum was designed to remove the power of local governments to include gay men, lesbians and bisexuals within the protection of their anti- discrimination laws. 25 Amendment 2 also would have cemented the anti-gay policy by requiring that future changes be made only by constitutional amendment rather than conventional legisla- 6 tion.2 The campaign for Colorado's Amendment 2 was waged largely on the theme of repealing "special rights" for homosexuals.2 7 Amendment 2's supporters put forth the proposition that anti-discrimination laws provided "special rights" for the listed groups, while suggesting that protection from discrimination was unavailable to anyone not listed. 28 The impression of "specialness" was reinforced by Amendment 2's references to "protected status," "minority status" and "quota preferences"-all code words evoca- tive of affirmative action.2 9 The "special rights" label allowed pro-

Discrimination in the Workplace: A Legal Reference Guide, 1 NAT'L J. SEXUAL ORIENTATION L. 33, 50-76 (1995) . 24. Jurisdictions in which similar amendments had been voted on or were pending include Arizona, California, Florida, Georgia, Idaho, Iowa, Maine, Michi- gan, Minnesota, Missouri, Montana, Ohio, Oregon and Washington. See Daniel J. Garfield, Don't Box Me In: The Unconstitutionality of Amendment 2 and English- Only Amendments, 89 NW. U. L. REV. 690, 691 n.7. 25. See Samuel A. Marcosson, The "Special Rights" Canard in the Debate over Lesbian and Gay Civil Rights, 9 NOTRE DAME J.L. ETHICS & PUB. POL'Y 137, 137 n.1 (1995). 26. See Richard F. Duncan & Gary L. Young, Homosexual Rights and Citizen Initiatives: Is Constitutionalism Unconstitutional?, 9 NOTRE DAME J.L. ETHICS & PUB. POL'Y 93, 96 (1995). 27. See Marcosson, supra note 25, at 138-39 nn.6-7 (pointing out the importance of the "special rights" label to strategy and electoral success of initiative campaigns). 28. See Hubert Morken, "No Special Rights": The Thinking Behind Colorado's Amendment #2 Strategy, Presented at Annual Meeting of the American Political Science Association 6 (Sept. 1-4, 1994) (on file with the Journal of Law & Inequal- ity: A Journal of Theory and Practice)(quoting Anton Marco, co-founder of Colo- rado for Family Values, the organization that initiated Amendment 2, as pointing out that "almost everyone gets fired sometime or other in their lifetime"). Cf. Duncan & Young, supra note 26, at 94 n.7 ("[Hlomosexuals already have the same rights everyone else has-the right to be protected against discrimination on the basis of their race, gender, religion, age, and disability."); Marcosson, supra note 25, at 140 ("For example, if gay men and lesbians were granted the right to commit murder, that would be a 'special right' in the sense that it is not possessed by heterosexuals."). 29. Amendment 2 provided: No Protected Status Based on Homosexual, Lesbian or Bisexual Orientation. Neither the State of Colorado, through any of its branches or departments, nor any of its agencies, political subdivisions, municipali- ties or school districts, shall enact, adopt or enforce any statute, regulation, ordinance or policy whereby homosexual, lesbian or bisexual orienta- 282 Law and Iequality [Vol. 15:275 ponents of Amendment 2 to maintain that they were not hostile to homosexuals, but only supportive of "fairness" to all by eliminating 30 "special rights" for some. This argument proved politically po- tent to many centrist voters who did not fall clearly into either the anti-gay or gay civil rights camps.3' Such voters can be described as opposed to discrimination in employment and public accommo- 32 dations, but hostile to anything perceived as "preferences." Opponents of gay civil rights made a number of other arguments in support of Amendment 2. For example, Amendment 2 was described as a cost-cutting measure, as important to the main- tenance of community order, as promoting family autonomy, as advancing religious liberty and as protective of children from psychological and physical harm.3 3 By the time the case reached the United States Supreme Court, however, the significance of some of these arguments had begun to wane among practitioners 34 and the

tion, conduct, practices or relationships shall constitute or otherwise be the basis of or entitle any person or class of persons to have or claim any minority status, quota preferences, protected status or claim of discrimination. This Section of the Constitution shall be in all respects self- executing. COLO. CONST. art. II, § 30b (emphasis added). 30. Anton Marco, co-founder of Colorado for Family Values, a leading propo- nent of Amendment 2, champions a similar approach called the "silent solution." See Morken, supra note 28, at 1. Marco would have government adopt a "neutral silence that refuses to grant homosexuals preferred status ...yet maintains equal protection for all persons." Id. at 7. The politically important "special rights" argument proved central in the ensuing litigation as well. Writing for the majority, Justice Kennedy identified it as Colorado's "principal argument." Romer v. Evans, 116 S. Ct. 1620, 1624 (1996). Justice Scalia incorporated much of the "special rights" rhetoric into his dissent. See, e.g., id. at 1630 ("The amendment prohibits the special treatment of homosexuals, and nothing more."); see also infra Part II.D. It is not difficult to foresee that the argument over "special rights" will endure as an important component of anti-discrimination cases. In objecting to the Su- preme Court's grant of certiorari in a case involving a city charter amendment similar to Colorado's constitutional amendment, Justice Scalia expressed his belief that "the people [may] decide, in democratic fashion, not to accord special protection to homosexuals." Equality Found. of Greater Cincinnati v. City of Cincinnati, 116 S. Ct. 2519 (1996) (Scalia, J., dissenting) (emphasis added). 31. See Marcosson, supra note 25, at 138 n.6. 32. Id; cf. Morken, supra note 28, at 7 ("Some moderates reject the whole civil rights emphasis on group rights and see voting yes on Amendment 2 as their opportunity to stop the further expansion of legal entitlements."). 33. See Evans v. Romer, 882 P.2d. 1335, 1339-40 (Colo. 1994), affd, 116 S. Ct. 1620 (1996). 34. Colorado did not assert the child protection rationale, arguably the most emotionally-charged argument, before the Court. See Evans, 882 P.2d. at 1340 n.2. The Court summarily rejected Colorado's other justifications-that Amend- ment 2 protected "other citizens' freedom of association" and "conserv[ed Colo- rado's] resources to fight discrimination against other groups"-saying, "[Tihe breadth of the Amendment is so far removed from these particular justifications that we find it impossible to credit them." Romer v. Evans, 116 S. Ct. 1620, 1629 (1996). 1997] ROMER V. EVANS public 35 alike. An argument that remained important involved the 36 position of homosexuals in society as a disadvantaged minority. Proponents of Amendment 2 argued that because homosexuals were economically "privileged" they did not need the protection of anti-discrimination laws.3 7 For example, the leading supporter of Amendment 2 argued that "gays are one of the most privileged groups in the country, sporting average annual household incomes over $55,000."38 In the context of the litigation, the fictitious eco-

35. Proponents of Amendment 2 may have been following the lead of Anton Marco. He believed that the "special rights" description of anti-discrimination laws justified the amendment far more effectively in the political arena than other arguments. See Morken, supra note 28, at 4, 9. Support for Amendment 2 de- clined when, against Marco's advice, the moral arguments based on disapproval of homosexuality were given greater prominence. Id. at 4. Moreover, Marco took the defeat of Oregon's similar (though more morally condemning) initiative as proof of his assessment's validity because proponents of Oregon's initiative utilized a morally-based strategy. Id. 36. See Brief for Petitioner at 42, Romer v. Evans, 116 S. Ct. 1620 (1996) (No. 94-1039) (defending Colorado's conclusion that gay men and lesbians are "politically powerful and relatively privileged"). The continued salience of this argument is understood as a rational response to prevailing equal protection doctrine. By casting gays and lesbians as privileged, proponents of Amendment 2 complicated the legal argument for their opponents because equal protection jurisprudence has for some time been influenced by Justice Harlan Fiske Stone's now famous "footnote 4" in Carolene Products. See United States v. Carolene Products Co., 304 U.S. 144, 152-53 n.4 (1938) (suggesting that judicial invalidation of legislative line-drawing is particularly justified where political action disadvantaging a "discrete and insular" minority renders the political process defective). Though minority status is no longer the only trigger to heightened scrutiny by the Su- preme Court, minority status, disadvantage and a history of discrimination still serve as prominent triggers to Court attention to classifications under equal protection. See infra notes 56-63 and accompanying text. 37. See Morken, supra note 28, at 6, 9 (discussing Marco's use of this argument). Cincinnati voters recently approved a city charter amendment similar to Colorado's Amendment 2 in its objective and wording. See Equality Found. of Greater Cincinnati, Inc. v. City of Cincinnati, 860 F. Supp. 417, 422 (1994), affd in part, vacated in part, 54 F.3d 261 (1995), cert. granted, 116 S. Ct. 2519 (1996). Pro- fessor James David Woodward, a political scientist at Clemson University, testi- fied in defense of the charter amendment that gays, lesbians and bisexuals were not politically powerless because of their affluence, education and intense feelings on certain political issues. Id. at 426. 38. See Jane S. Schacter, The Gay Civil Rights Debate in the States: Decoding the Discourse of Equivalents, 29 HARV. C.R.-C.L. L. REV. 283, 292 (1994) (quoting Will Perkins, Executive Chairman of the Board of Colorado for Family Values). But see Eric Heinze, Gay and Poor, 38 HOW. L.J. 433 (1995) (asserting that "gays are economically disadvantaged in the United States due to "direct" and "indirect" discrimination"); cf. M.V. Lee Badgett, The Wage Effects of Sexual Orientation Dis- crimination,INDUS. & LAB. REL. REV. 726 (1995) (finding significant wage dispar- ity between homosexual and heterosexual men of similar education, experience and occupation, and also finding that lesbian and bisexual women earned less than their heterosexual counterparts). Law and Inequality [Vol. 15:275 nomic well-being of homosexuals was used to argue that judicial 39 protection was unjustified. Once Amendment 2 moved to litigation, the fact that it had passed by referendum 40 allowed anti-gay forces to emphasize the right of the people to govern themselves. Referendum supporters argued that gay civil rights policy was a matter that rightly be- 4 1 longed to the "people," not to a court of unelected officials. Amendment 2's proponents hoped to persuade the Court to defer to the judgment of the Colorado electorate. 42 The immediate public reaction suggested that at least part of this populist argument 43 had some influence. Another argument likely to be important in the larger political debate for similar reasons is Justice Scalia's claim, made in the dissent, that the Court has taken sides in a "culture war."44 Court interventions aimed at assuring an open and fair political process carry a much greater legitimacy than court actions which seem to 45 take a position on substantive policy.

39. See Brief for Petitioner at 42 n.30, Romer (No. 94-1039). 40. Amendment 2 passed by a vote of 813,966 (53.4%) to 710,151 (46.6%) on November 3, 1992. Evans v. Romer, 854 P.2d 1270, 1272 (Colo. 1993). 41. See Duncan & Young, supra, note 26, at 97. "Grassroots initiatives like Amendment Two are the product of a populist revolt against restrictive homosexual rights laws." Id. at 134. See also Morken, supra note 28, at 7 (arguing for democratic legal reform rather than "government intrusion into what ought to be left to private opinion and the negotiations of the market"). 42. See Brief for Petitioner at 7-8, 10, Romer (No. 94-1039). "[Tlhe people approved Amendment 2 after a spirited campaign which the trial court found to be a classic example of democratic self-government." Id. at 7 (emphasis added). "The reasoning of the [Colorado Supreme Court] does profound violence to settled un- derstandings of the authority of, and respect for, popular government at the state level." Id. at 10 (emphasis added). 43. Cf. supra note 17 (discussing the reaction of a Northwestern University School of Law professor). 44. See Romer v. Evans, 116 S. Ct. 1620, 1637 (1996). 45. Court actions that defend democracy have the greatest legitimacy. Studies repeatedly conclude that the Supreme Court is most likely to be followed when its decision taps into a long held public value, even if the value has not been implemented. For example, one commentator attributed the ultimate ability of Brown v. Board of Education, 347 U.S. 483 (1954), to persuade Congress to join in effect- ing social change to the fact that "Brown carried with it the moral conscience of the American people." HARRY P. STUMPF, AMERICAN JUDICIAL POLITICS 469 (1988). Similarly, David M. O'Brien, a well-known scholar of the Court, has observed: "At its best, the Court appeals to the country to respect the substantive value choices of human dignity and self-governance embedded in our written Constitution." DAVID M. O'BRIEN, STORM CENTER: THE SUPREME COURT IN AMERICAN POLITICS 321 (1986). Much of the debate surrounding proper modes of constitutional interpretation starts from the proposition that the Court's power becomes suspect when it questions the actions of the majority. See ALEXANDER M. BICKEL, THE LEAST 1997] ROMER V. EVANS

The "special rights" and governance arguments put forth were for both public consumption and to persuade the Court of the constitutionality of Amendment 2.46 The legal argument, however, was for the Court alone; it was, and had to be, more complicated because it needed to address the prevailing standards under the equal protection doctrine. For example, proponents of Amendment 2 needed to argue the proper standard of review, a matter little understood within, and of little relevance for, the public debate. A discussion of these arguments follows.

I. The Decision

A. ConstitutionalEnvironment The core legal issue in Romer was whether Amendment 2 violated the Equal Protection Clause of the Fourteenth Amend- ment. The Equal Protection Clause provides that "No State shall .. deny to any person within its jurisdiction the equal protection of the laws." 47 The core principle of the Equal Protection Clause is, as Justice Kennedy stated in Romer, a "commitment to the law's neutrality where the rights of persons are at stake."48 This means that government must treat those who are similarly situated the same.49 There are two basic ways in which this constitutional pro-

DANGEROUS BRANCH: THE SUPREME COURT AT THE BAR OF POLITICS 16-23 (2d ed. 1986) (describing this problem as the "countermajoritarian difficulty"). John Hart Ely's theory of representation-reinforcement was designed to resolve this difficulty by confining legitimate Court decision-making to correcting defects in the democratic process. See JOHN HART ELY, DEMOCRACY AND DISTRUST: A THEORY OF JUDICIAL REVIEW 77-80, 101-04, 135-79 (1980). For a defense of Ely's position, see Michael J. Klarman, The Puzzling Resistance to PoliticalProcess Theory, 77 VA. L. REV. 747, 768-819 (1991). Some of Ely's critics would accept representation-reinforcement as legitimate only if coupled with a theory protective of fundamental liberties. See, e.g., Lau- rence H. Tribe, The Puzzling Persistenceof Process-Based ConstitutionalTheories, 89 YALE L.J. 1063, 1072-77 (1980). The view that Court action removing defects in the democratic process is especially justified originated in Justice Harlan Fiske Stone's footnote 4 in United States v. Carolene Products Co., 304 U.S. 144 (1938). See Klarman, supra, at 749 (calling the footnote "the seminal statement of political process theory"); see also supra note 36 (discussing the now famous footnote 4 in more detail). 46. See supra notes 27-28, 30 and accompanying text. 47. U.S. CONST. amend. XIV, § I. 48. Romer, 116 S. Ct. at 1623. 49. RONALD D. ROTUNDA & JOHN E. NOwAK, 3 TREATISE ON CONSTITUTIONAL LAW: SUBSTANCE AND PROCEDURE § 18.2, at 7 (2d ed. 1992). This formulation of the Fourteenth Amendment principle and the similar equal protection component of the Fifth Amendment appears, with some minor wording variations, in several cases. See, e.g., F.S. Royster Guano Co. v. Virginia, 253 U.S. 412, 415 (1920) ("[Tihe classification must be reasonable, not arbitrary, and must rest upon some Law and Inequality [Vol. 15:275 tection can be violated.50 First, "equality can be denied when government classifies so as to distinguish, in its rules or programs, between persons who should be regarded as similarly situated in terms of the relevant equal protection principles."5 1 In addition, "equality can be denied when government fails to classify with the result that its rules or programs do not distinguish between persons who, for equal protection purposes, should be regarded as dif- 52 ferently situated." There often is difficulty, however, in determining whether an apparent difference in situations is meaningful or not. Problems also arise in attempting to draw lines with such precision that only those-and all those-who share the characteristics important to the achievement of the government's ends will be singled out by legislation.53 Because of these difficulties, a legislature must be given broad latitude if it is to successfully assess a social problem, weigh competing concerns and remedy the ill.54 Thus, if government is to function effectively it must classify citizens even though the unavoidable result is that some individuals will be treated less 55 favorably than others. ground of difference having a fair and substantial relation to the object of the legislation, so that all persons similarly circumstanced shall be treated alike."); United States Dep't of Agric. v. Murry, 413 U.S. 508, 517 (1973) (Marshall, J., concurring) (arguing that the Due Process Clause of the Fifth Amendment requires that "individuals similarly situated must receive the same treatment by the Gov- ernment"). 50. LAURENCE H. TRIBE, AMERICAN CONSTITUTIONAL LAW § 16-1, at 1438 (2d ed. 1988). 51. Id. 52. Id. 53. See GEOFFREY R. STONE ET AL., CONSTITUTIONAL LAw 539 (2d ed. 1991). The author describes the equal protection difficulties inherent in statutory con- struction as follows: For almost in all cases, the classifications will not be perfectly efficient, but will be either "overinclusive" or "underinclusive" or both. A classification is "overinclusive" if it disadvantages some people who do not in fact threaten the state's interest. It is "underinclusive" if some people are not disadvantaged even though they threaten the state's interest. Id. The legislative determination of the age at which one can drive a car provides a good example. Presumably, the primary objective is to foster traffic safety. Whatever the age set, there will be some individuals who would be safe drivers although they are younger than the set age and there will be some individuals over the set age who are unsafe drivers. 54. Plyler v. Doe, 457 U.S. 202, 216 ("A legislature must have substantial latitude to establish classifications that roughly approximate the nature of the problem perceived, that accommodate competing concerns both public and private, and that account for limitations on the practical ability of the State to remedy every ill."). 55. See Dandridge v. Williams, 397 U.S. 471, 485 (1970) (arguing that the practical problems of effective governing may justify a classification even if in practice it results in some inequality). 1997] ROMER V. EVANS

To address the need "to reconcile the principle with the real- 56 ity" as well as the need to minimize the Court's intrusion into the political process, the Court has developed an analytic framework for assessing claims under the Equal Protection Clause. Until the 1940s, the Equal Protection Clause was effectively invoked only as a protection against discrimination based on race or national origin. 57 From the 1940s to the 1970s, the Court employed a two- tiered analytic approach. Most classifications were upheld as long as they were reasonable, and the Court deferred to the legislative or executive judgment as to what was reasonable.58 Government action assessed under this "rational basis" test was permissible if there was any conceivable basis for it.9 Most classifications were reviewed under this rationality standard; under such deferential review, invalidation of legislation was rare. 60 However, no similar

56. Romer v. Evans, 116 S. Ct. 1620, 1627 (1996). 57. The first case to invalidate a racial classification was Strauder v. West Vir- ginia, 100 U.S. 303 (1880) (invalidating a West Virginia law that limited jury duty to white males), abrogated by Taylor v. Louisiana, 419 U.S. 522 (1975). Shortly thereafter, the Court held that the Equal Protection Clause protected Chinese aliens residing in San Francisco from discrimination in the administration of the law. See Yick Wo v. Hopkins, 118 U.S. 356 (1886) (finding Yick Wo's imprisonment for violation of a San Francisco ordinance barring laundries in wooden buildings without a variance to be invalid because only Chinese were denied such variances). At the same time, however, the Court often allowed classifications based on race and national origin that were disadvantageous to minorities. The best known example of this is Plessy v. Ferguson, 163 U.S. 537 (1896) (upholding a Lousiana law that required the separation of the races on railroad cars), overruled by Brown v. Board of Educ., 347 U.S. 483 (1954). Plessy set into motion the "separate but equal" era, a period in which the Court's view of the equality prong was often strained. See, e.g., Cumming v. Richmond County Bd. of Educ., 175 U.S. 528 (1899) (finding no equal protection violation in a decision by Richmond County, Georgia to fund a white but not a black high school, thereby denying black school children an opportunity to receive a high school education). 58. See Railway Express Agency v. New York, 336 U.S. 106 (1949) (upholding a New York traffic regulation prohibiting advertising on business vehicles but permitting an exception for vehicles operated by the owner of the business being ad- vertised). 59. Williamson v. Lee Optical of Okla., Inc., 348 U.S. 483 (1955) (upholding a law that required opticians, but not sellers of ready-to-wear glasses, to obtain a prescription before replacing a customer's glasses); McGowen v. Maryland, 366 U.S. 420 (1961) (upholding, with an extremely deferential review, exemptions for some businesses from Sunday closing laws). The Court has not always used such deferential language, however. In F.S. Royster Guano Co. v. Virginia, 253 U.S. 412, 415 (1920), the Court asserted that to be rational the classification had to have a "fair and substantial relationship to the object of the legislation." Id. (upholding a Virginia tax law that distinguished between businesses that had some dealings within the state and those whose operations were conducted wholly outside the state). 60. One of the exceptions was Morey v. Doud, 354 U.S. 457 (1957) (invalidating an Illinois statute regulating "currency exchanges" but exempting the American Express Company), overruled by City of New Orleans v. Dukes, 427 U.S. 297 288 Law and Inequality [Vol. 15:275 presumption of constitutionality existed when the government 61 used a classification that the Court had designated as "suspect" or when government action affected the exercise of a "fundamental right."62 In those circumstances, the Court has required the State to demonstrate that it has a compelling interest in the classifica-

(1976). See also Metropolitan Life Ins. Co. v. Ward, 470 U.S. 869 (1985) (striking down an Alabama tax scheme that favored its own residents). 61. The notion that some types of classifications may be "suspect" and require heightened scrutiny has its origin in Korematsu v. United States, 323 U.S. 214 (1994). Writing for the Court, Justice Black stated that "all legal restrictions which curtail the civil rights of a single racial group are immediately suspect. That is not to say that all such restrictions are unconstitutional. It is to say that courts must subject them to the most rigid scrutiny." Id. at 216. The Court went on, however, to uphold the removal of Japanese-Americans from designated mili- tary zones without requiring the evidence that strict scrutiny would seem to require. Id. at 219. Since Korematsu, the Court has been more rigorous in applying the strict scrutiny standard, invalidating almost all classifications that come under it. See, e.g., Palmore v. Sidoti, 466 U.S. 429 (1984) (invalidating a Florida Court order awarding custody to the father of a child when the mother became involved in an interracial marriage); Loving v. Virginia, 388 U.S. 1 (1967) (striking down Vir- ginia's laws against interracial marriage). 62. The first such decision was Skinner v. Oklahoma, 316 U.S. 535, 541 (1942) (striking down on equal protection grounds Oklahoma's Habitual Criminal Sterili- zation Act, which required that persons convicted two or more times of felonies be sterilized but exempted those convicted of embezzlement and other "white collar" crimes from such punishment). Like the right to procreate, access to the ballot was deemed a fundamental interest warranting heightened scrutiny in later cases. See City of Phoenix v. Kolodziejski, 399 U.S. 204 (1970) (invalidating election where voting rights were limited to property owners); Cipriano v. City of Houma, 395 U.S. 701 (1969) (striking down a Louisiana law limiting voting rights to property owners); Kramer v. Union Free Sch. Dist. No. 15, 395 U.S. 621 (1969) (striking down a New York law limiting voting in school board elections to parents or property owners); Harper v. Virginia Bd. of Electors, 383 U.S. 663 (1966) (striking down a poll tax). But see Salyer Land Co. v. Tulare Lake Basin Water Storage Dist., 410 U.S. 719 (1973) (upholding a statutory scheme limiting voting rights in water storage districts to landowners). The right to travel interstate was recognized as fundamental in Shapiro v. Thompson, 394 U.S. 618 (1969) (striking down various state statutes prohibiting welfare benefits to residents of less than one year as violative of their right to interstate movement), overruled on other grounds by Edelman v. Jordan, 416 U.S. 1000 (1974). The right to travel interstate played a part along with concern for ballot access in the invalidation of a Tennessee year-long residency requirement for voting in Dunn v. Blurnstein, 405 U.S. 330 (1972). Access to the courts has also been treated as fundamental in some circumstances. See Griffin v. Illinois, 351 U.S. 12 (1956) (holding that indigent criminal defendants can not be denied appellate review allowed other defendants because they lack the means to pay for a transcript); Boddie v. Connecticut, 401 U.S. 371 (1971) (invalidating Connecticut's refusal to waive court costs for indigent couples seeking a divorce because the interest involved, i.e., marriage, is fundamental). As recently as 1996, the Court struck down a Mississippi requirement making the appeal of a judgment terminating parental rights contingent upon the advance pay- ment of $2,352.36. See M.L.B. v. S.L.J., 117 S. Ct. 555 (1996). 1997] ROMER V. EVANS 289 tion, and that the means it selected were narrowly tailored to ef- 63 fect that interest. If the Court still used this two-tiered approach, the plaintiffs in Romer would have had to convince the Court either that sexual orientation should be considered a suspect classification or that Amendment 2 limited a fundamental right. There were three reasons why it was unlikely that the Romer Court would add sexual orientation to the suspect classification category. First, the Court had already indicated its great reluctance to broaden the suspect 64 classification category in its analysis of gender and illegitimacy. Second, such a holding might seem to be in direct conflict with the Court's ruling in Bowers v. Hardwick,65 although that case turned on the Due Process Clause.66 Third and most importantly, desig- nating sexual orientation a suspect classification would transform laws throughout the nation and surely cause political turmoil. Such a course was not one this fairly cautious Court was likely to take.67 Foregoing a suspect classification argument and instead convincing the Court that Amendment 2 encroached on a funda- 68 mental right seemed the more promising approach. However, this two-tiered analysis has given way to a more complicated system. In the 1970s, the Court created an intermediate level of review. Originating with gender classifications, the

63. See, e.g., In re Griffiths, 413 U.S. 717, 721-22 (1973). The strict scrutiny standard as been described as "'strict' in theory and fatal in fact." Gerald Gunther, The Supreme Court: A Model for Newer Equal Protection, 86 HARV. L. REV. 1, 8 (1972). 64. This reluctance with regard to illegitimates can be seen in Labine v Vin- cent, 401 U.S. 532 (1971) (upholding state intestate succession statutes that favored legitimate over illegitimate children). Rather broaden the suspect classification category, the Court has applied an intermediate level of review to these classifications. This new standard, which the Court has described as falling somewhere between strict scrutiny and rational basis review, was openly acknowledged in Craig v. Boren, 429 U.S. 190, 197 (1976) (holding that a statute that classifies based on gender "must serve important governmental objectives and must be substantially related to achievement of those objectives"). Although not recognized until Craig, the standard was actually used in earlier decisions such as Reed v. Reed, 404 U.S. 71, 76 (1971) (invalidating Idaho probate code provision that preferences men over women in appointment as administrator of an estate). 65. 478 U.S. 186 (1986). 66. In Bowers, the Court held that a Georgia law criminalizing sodomy did not violate the Due Process Clause of the Fourteenth Amendment. Though the law applied to all sodomy, the majority focused on whether the right of privacy extended to provide homosexuals with a fundamental right to engage in gay sex. Id. at 190. 67. See infra notes 190-202 and accompanying text; see also TRIBE, supra note 50, § 16-3, at 1445. 68. A fundamental rights argument did persuade the trial court and the Colo- rado Supreme Court. See infra notes 80-87 and accompanying text. Law and Inequality [Vol. 15:275 notion has been extended to other categories. 69 For classifications falling within this intermediate category, government must demonstrate an "important" interest and a "substantial" relationship of 70 the classification to the interest. Moreover, rational basis review itself has been refined. No longer does the rational basis test automatically result in the vali- dation of the government's action. 71 There have been circumstances, although not clearly defined, in which the Court has been less willing than it once was to credit the legitimacy of government purposes.7 2 Obviously, whatever the framework, the "special rights" controversy in Romer was pivotal; 73 if Amendment 2 had been viewed as not treating homosexuals unfavorably, then none of the equal protection criteria would have applied.

B. ProceduralHistory Colorado voters adopted Amendment 2 in November 1992.74 Shortly thereafter Richard Evans, along with several other individuals and adversely affected local governmental entities, peti- tioned the Denver district court for an order declaring Amendment 7 5 2 unconstitutional and an injunction barring its enforcement.

69. TRIBE, supra note 50, § 16-33, at 1610. 70. See Craig v. Boren, 429 U.S. at 197 (striking down an Oklahoma statute barring the sale of 3.2% beer to males, but not to females, under the age of twenty- one); Trimble v. Gordon, 430 U.S. 762 (1977) (striking down an Illinois law that prevented children born out of wedlock from inheriting from their fathers unless paternity had been acknowledged in a judicial proceeding prior to the father's death). 71. See, e.g., United States Dep't of Agric. v. Moreno, 413 U.S. 528, 534 (1973) (holding that unrelated persons occupying one household could not be denied food stamps). 72. Compare Railway Express Agency v. New York, 336 U.S. 106 (1949) (where the court strongly deferred to a New York traffic regulation prohibiting advertising on business vehicles but making an exception for vehicles operated in the business of the owner), with City of Cleburne v. Cleburne Living Ctr., 473 U.S. 432 (1985) (holding the City's denial of a use permit for a group home for mentally retarded. individuals violated the equal protection clause). For a further discussion of the Court's enhanced rational basis test, see Part III.A. 73. For a discussion of the "special rights" controversy, see infra Part III.C. 74. Romer v. Evans, 116 S. Ct. at 1620, 1623 (1996). The cities of Aspen, Boul- der and Denver had ordinances that protected gays and lesbians which were con- sequently repealed by Amendment 2. Id. Amendment 2 repealed all laws and policies barring discrimination of homosexuals and forbidding the adoption of "all legislative, executive or judicial action at any level of state or local government designed to protect the named class." Id. 75. See Evans v. Romer, 854 P.2d 1270, 1272 (Colo. 1993) (Evans I) (explaining the plaintiffs' request that the court "enjoin the enforcement of Amendment 2" and declare it to be unconstitutional), affd, 882 P.2d 1335 (Colo. 1994), affd, 116 S. Ct. 1620 (1996). 1997] ROMER V. EVANS

As already suggested, Evans first had to convince the Court that there had been discrimination against gay men, lesbians and bisexuals. Because the more traditional and widely used rational basis test generally results in deference to voters and the legislature, Evans and his co-plaintiffs next had to persuade the Court to 76 apply some form of heightened scrutiny. At the Denver district court, Evans argued that a classification based on sexual orientation should be considered suspect or quasi-suspect.77 Alternatively, Evans argued that the court should use rational basis review and that "the state's purported compelling interests are not rationally related to the enactment of Amendment 2."78 The trial court refused to place sexual orientation in the suspect category; 79 however, it still held that strict scrutiny was warranted because Amendment 2 burdened the fundamental right "not to have the State endorse and give effect to private biases."80 The trial court concluded that Amendment 2 must be reviewed under the strict scrutiny standard.8 1 Because there was a reasonable probability plaintiffs would prevail under that standard, the district court issued a preliminary injunction preventing enforcement of Amendment 2 pending a trial on the 8 3 merits.8 2 Colorado appealed.

76. See supra notes 56-63 and accompanying text (discussing rational basis review as a traditionally deferential standard); but see supra notes 69-70 and accompanying text (explaining that rational basis no longer automatically results in deference to the government). 77. Evans v. Romer, 882 P.2d 1335, 1341 n.3 (Colo. 1994) (Evans II), affd, 116 S. Ct. 1620 (1996). 78. Id. at 1341; see also Evans 1, 854 P.2d at 1273. 79. Evans II, 882 P.2d at 1341 n.3 (noting that the trial court rejected the plaintiffs' argument that sexual orientation is a suspect class because that group is not politically powerless). 80. Evans 1, 854 P.2d at 1274. 81. Id. The court stated, "A legislative enactment which infringes on a fundamental right or which burdens a suspect class is constitutionally permissible only if it is 'necessary to promote a compelling state interest'... and does so in the least restrictive manner possible." Evans II, 882 P.2d at 1341 (citing Dunn v. Blum- stein, 405 U.S. 330, 342 (1972)). Under prevailing Equal Protection Clause jurisprudence, the Court looks more closely at some classifications than at others. See supra notes 55, 59 and accompanying text (identifying suspect classifications). Under strict scrutiny, the state must show a compelling interest and a means closely tailored to that interest. See supra notes 59-61 and accompanying text (explaining strict scrutiny and the burden of proof it imposes on the government). Strict scrutiny is applied if the classification is one the Court has designated as suspect or if the interest regulated is fundamental. See supra notes 59-60 and accompanying text (explaining when strict scrutiny applies). 82. Evans I, 854 P.2d at 1273. 83. Id. at 1274. Law and Inequality [Vol. 15:275

The Colorado Supreme Court's first review of the case (Evans 1) focused on the standard of review. The state supreme court agreed with the trial court's finding that Amendment 2 implicated a fundamental right and even recharacterized it as a fundamental right to have an equal voice in government.84 The supreme court remanded the case to the district court "to determine whether [Amendment 2] is constitutionally valid under the Equal Protec- tion Clause."8 5 On remand, the trial court held that the State 8 6 failed to meet its heavy burden under the strict scrutiny test. Colorado offered a number of additional justifications for Amend- ment 2, each of which the trial court dismissed as either not com- 8 7 pelling or not necessary to the interest at stake. On appeal, the Colorado Supreme Court affirmed its Evans I holding regarding the standard of review.88 It also agreed with the trial court's disposition on the merits. Although acknowledging that some of the interests Colorado advanced were compelling, the Colorado Supreme Court agreed with the trial court that Amend- ment 2 was not narrowly tailored to effect those ends.8 9 Thus, while the supreme court recognized that the State had a compelling interest in religious liberty, it reasoned that "an equally effective, and substantially less onerous way of accomplishing that purpose simply would be to require that anti-discrimination laws

84. Id. at 1276. 85. Id. at 1286. 86. See Evans v. Romer, 882 P.2d 1335, 1340-41 (Colo. 1994), affd, 116 S. Ct. 1620 (1996). 87. Id. at 1339-40. Colorado asserted the following compelling interests: (1) deterring factionalism; (2) preserving the integrity of the state's political functions; (3) preserving the ability of the state to remedy discrimination against suspect classes; (4) preventing the government from interfering with personal, familial, and religious privacy; (5) preventing government from subsidizing the political objectives of a special interest group; and (6) promoting the physical and psychological well-being of Colorado children. Id. at 1339. The trial court held these either not compelling or not narrowly tailored to achieve the purported objectives. Id. 88. Id. at 1341. 89. The Colorado Supreme Court reviewed each of the State's claims de twvo. Id. at 1341-50. It agreed with the trial court's position that the State's asserted interests were either insufficiently compelling or were achievable by less intrusive means. See id. (detailing which state interests were insufficiently compelling or too intrusive). For example, the asserted interest in religious freedom was compelling but was not the least restrictive means because religious freedom could be achieved by providing an exemption based on religious convictions. See id. at 1340-41. 1997] ROMER V. EVANS which include provisions for sexual orientation also include exceptions for religiously-based objections." 90 Similarly, the Colorado Supreme Court noted that although the State might have a compelling interest in protecting family privacy, the interest was not implicated in the case at hand.91 Moreover, such an interest would not extend to having the State endorse one set of moral values over another. 92 Additionally, the court was not convinced that Amendment 2 was justified as a necessary means for protecting associational privacy.93 Unmoved by these arguments and others,94 the Colorado Supreme Court dismissed the State's justifications as not compelling.95

C. The U.S. Supreme Court Opinion Justice Kennedy, writing for the United States Supreme Court, began his remarks in Romer v. Evans by invoking Justice John Harlan's dissent in Plessy v. Ferguson-'-"the Constitution neither knows nor tolerates classes among citizens"96 -followers of the case must have anticipated that Evans would win. Continu- ing, Justice Kennedy explained that these words signify a "commitment to the law's neutrality where the rights of persons are at stake."97 However, those who thought the reference meant that sexual orientation would be treated as a suspect or quasi- suspect classification requiring heightened scrutiny were mis- taken. The U.S. Supreme Court affirmed the Colorado Supreme Court's result, but it did so on quite different grounds. Whereas the Colorado Supreme Court had subjected Amendment 2 to a searching review, which Amendment 2 failed, the U.S. Supreme Court not only held that Amendment 2 failed constitutional analy-

90. Id. at 1343. 91. Id. at 1343-44. 92. Id. at 1343. 93. Id. at 1344-45. The Colorado Supreme Court noted that Amendment 2 prohibited protection of gays and lesbians in activities that had little or no connection to intimate relationships. Id. at 1345. 94. The court stated that "the preservation of fiscal resources, administrative convenience, and the reduction of the workload of governmental bodies" were not compelling interests. Id. 95. Id. 96. Romer v. Evans, 116 S. Ct. 1620, 1623 (1996) (quoting Plessy v. Ferguson, 163 U.S. 537, 559 (1896) (Harlan, J., dissenting) (upholding a Louisiana law that required the separation of races on railroad cars), overruled by Brown v. Board of Educ., 347 U.S. 483 (1954)). 97. Id. Law and Inequality [Vol. 15:275 sis 98 but also that it "was not within our constitutional tradition to enact" the law.99 Whereas the state supreme court agreed that some of Colorado's justifications were compelling even though it held that they could be advanced less intrusively, 0 0 the U.S. Su- preme Court did not take the State's justifications seriously because they were too far removed from Amendment 2 itself.1 1 Section I of the U.S. Supreme Court's opinion provided a straightforward account of the law's provisions and the history of the litigation. 0 2 The substantive analysis began in Section II with the consideration and rejection of Colorado's "special rights" claim.'0 3 The Court stated that the State's assertion that Amend- ment 2 did no more than put homosexuals "in the same position as all other persons" was "implausible."'' 0 4 There was the undeniable fact that Amendment 2 placed gays, lesbians and bisexuals in a less favorable position than they had been before its adoption. 0 5 Moreover, the harm imposed was extensive, as documented by the state supreme court. 10 6 The Court then explained that civil rights laws cannot be viewed as providing "special" advantages to those covered because in most cases, people do not face arbitrary exclusions from every- day activities. 0 7 This presumption of non-discrimination is re- flected in the common law rule that "innkeepers, smiths, and others who 'made profession of a public employment,' were prohibited from refusing, without good reason, to serve a customer."'' 0 While this rule might have served the majority well enough, it was often insufficient to protect some identifiable and unpopular groups. 10 9 Civil rights laws were designed to cope with this sort of group disadvantaging.1n °

98. Id. at 1627. 99. Id. at 1628. 100. See supra notes 87-89 and accompanying text. 101. 116 S. Ct. at 1629. 102. Id. at 1623-24. 103. Id. at 1624-25. The State's "special rights" claim asserted that Amendment 2 "does no more than deny homosexuals special rights." Id. at 1624. 104. Id. 105. Id. at 1625. The Court explained that the "amendment withdraws from homosexuals, but no others, specific legal protection from the injuries caused by discrimination." Id. 106. Id. 107. Id. at 1627. 108. Id. at 1625 (quoting Hurley v. Irish-American Gay, Lesbian and Bisexual Group of Boston, Inc., 515 U.S. 115 S. Ct. 2338, 2346 (1995)). 109. Id. 110. See id. (explaining that Colorado's state and municipal laws "enumerate 1997] ROMER V. EVANS

Furthermore, Justice Kennedy noted that Amendment 2 also repealed regulations forbidding discrimination based on sexual orientation in public employment."1 Even though he did not explain the importance of this point, the argument responded to Colorado's claim, implicit in the "special rights" argument, that Amendment 2 incorrectly placed the right to be free from discrimination ahead of the liberty rights of those who wished to discriminate."1 2 Finally, Justice Kennedy wrote that Amendment 2 denied homosexuals the "safeguards others enjoy."" 3 Moreover, unlike others gay men, lesbians and bisexuals could only secure equal rights by amending the state constitution or "by trying to pass helpful laws of general applicability;"" 4 that is, laws barring the arbitrary disadvantaging of any individual. Having argued that Amendment 2 did indeed harm homosexuals, the Court then turned its attention to the case's second principal issue: whether such harm is barred by the Fourteenth Amendment. 115 It is worth repeating that notwithstanding Justice Kennedy's opening remarks, the Court has never required the government to act with perfect neutrality." 6 As Justice Kennedy acknowledged, if government is to function effectively, it must classify citizens, even though the unavoidable, result is that some groups will be treated less favorably than others. 117 The need "to the groups or persons" which the laws protect and that these groups are not limited to those which receive heightened equal protection scrutiny in the courts). 111. Id. at 1626. The Court explained the breadth of Amendment 2 by stating that it applies to public employment, public accommodations and various private transactions. Id. 112. See Evans v. Romer, 882 P.2d 1335, 1342 (Colo. 1994), affd, 116 S. Ct. 1620 (1996) (explaining the defendants' argument that private citizens, such as land- lords, would have to unfairly compromise their "deep-seated and profound religious objections to homosexuality" in order to comply with the amendment). 113. Romer, 116 S. Ct. at 1627. Justice Kennedy summed up the Court's position as follows: We find nothing special in the protections Amendment 2 withholds. These are protections taken for granted by most people either because they already have them or do not need them; these are protections against exclusion from an almost limitless number of transactions and endeavors that constitute ordinary civic life in a free society. Id. 114. Id. 115: Id. 116. See discussion supra notes 47-55 and accompanying text (discussing judicial review of equal protection claims and explaining that if the government satis- fies the applicable constitutional standard, then it may use a given classification, even if that classification has uneven or non-neutral effects). 117. Romer, 116 S. Ct. at 1627. The Court explained that while the requirement of a "link between classification and objective" guides and limits government action, ordinarily a "law will be sustained if it can be said to advance a legitimate government interest, even if the law ... works to the disadvantage of a particular Law and Inequality [Vol. 15:275 reconcile the principle with the reality""l 8 led the Court to develop the two-tiered analytic approach described previously: If a law either burdens a fundamental right or targets a suspect class then the Court should examine it using strict scrutiny; if not, then the Court need only ensure that the classification "bears a rational relationship to some legitimate end."119 However, given the exceptional nature of Amendment 2, the Court perceived no need to utilize its traditional two-tiered analytic framework nor did it need to explicitly invoke either strict scrutiny or a rational basis test. Instead, the Court implicitly drew elements from each. It was important that Amendment 2 affected participation in the political process, but the Court's decision was not grounded in the fundamental rights component of equal protection analysis. In fact, if it had been, the Court indicated that it might go further than had the Colorado Supreme Court. Whereas the state supreme court took the position that a law limiting the exercise of one's "fundamental right to participate equally in the political process" would be invalid unless justified by a compelling interest and by a means necessary to promote that interest, 120 Justice Kennedy seemed to suggest that such a law was never permissible. 121 Although it was also important that the law was aimed at an identifiable and unpopular group, 122 the Court avoided holding that sexual orientation was a suspect classification calling for strict scrutiny. 123 Finally, the Court considered the 124 breadth of the inflicted harm to be an important factor. The Court noted that under ordinary circumstances, "a law will be sustained if it can be said to advance a legitimate government interest, even if the law seems unwise or works to the disadvantage of a particular group, or if the rationale for it seems tenu- 125 ous."' However, even deferential analysis implies the pursuit of group." Id. 118. Id. 119. Id. See also discussion supra notes 58-63 and accompanying text. 120. Evans v. Romer, 854 P.2d 1270, 1276 (Colo. 1993), affd, 882 P.2d 1335 (Colo. 1994), affd, 116 S. Ct. 1620 (1996). 121. See Romer, 116 S. Ct. at 1628. In writing for the majority, Justice Kennedy remarked that a law which makes it "more difficult for one group of citizens than for all others to seek aid from the government is itself a denial of equal protection of the laws in the most literal sense." Id. 122. See id. at 1628. 123. Id. at 1629. Rather than imposing strict scrutiny, the Court only applied rational basis review to Amendment 2, and it held that the law was not "directed to any identifiable legitimate purpose or discrete objective." Id. 124. See id. Completely denying protection of the laws to a group of persons based upon one trait is "unprecedented in our jurisprudence." Id. 125. Id. at 1627 (citation omitted). 1997] ROMER V. EVANS a legitimate end. 126 Moreover, the state may not be partial for its own sake, and it may not pass a law "for the purpose of disadvantaging the group burdened by the law."'127 Finally, the Court asserted that equal protection means that "a bare desire to harm a politically unpopular group cannot constitute a legitimate govern- 128 mental interest."' The Court then concluded that Colorado had pursued an illegitimate end.129 "[H]omosexuals ...are put in a solitary class with respect to transactions and relations in both the private and governmental spheres. The amendment withdraws from homosexuals, but no others, specific legal protection from the injuries caused by discrimination, and it forbids reinstatement of these laws and policies."'130 Furthermore, Colorado created this "solitary class" not only without adequate justification but without any credible justification. The Court could not take the State's justifications seriously given the undifferentiated, broad reach of the Amend- ment. 13' Therefore, Colorado failed to show that it had a proper purpose. "We must conclude that Amendment 2 classifies homosexuals not to further a proper legislative end but to make them 32 unequal to everyone else."' One final aspect of Justice Kennedy's opinion warrants mention. As the Court gave only passing attention to precedent, its ruling was vulnerable to Justice Scalia's charge that it was without foundation in the law. 3 3 One exception was the Court's effort to distinguish Romer from cases such as Williamson v. Lee Opti- cal' 34 and Railway Express Agency v. New York' 35-- decisions which upheld classifications that were at best loosely connected to the state's objectives. Justice Kennedy distinguished the classify-

126. See supra note 72 and accompanying text. 127. See Roiner, 116 S. Ct. at 1627 (citation omitted). 128. Id. at 1628 (quoting United States Dep't of Agric. v. Moreno, 413 U.S. 528, 534 (1973)). 129. Romer, 116 S. Ct. at 1629. 130. Id. at 1625. 131. "The breadth of the Amendment is so far removed from these particular justifications that we find it impossible to credit them." Id. at 1629. 132. Id. 133. Id. at 1633 (Scalia, J., dissenting). 134. 348 U.S. 483 (1955) (upholding a state law that discriminated between opticians and licensed ophthalmologists or optometrists based on the supposition that the regulation bore a rational relationship to public health). 135. 336 U.S. 106 (1949) (upholding a state law that discriminated between owners and non-owners when regulating advertisements on the sides of trucks in the name of traffic safety). Law and Inequality [Vol. 15:275 ing statutes challenged in these cases on the basis of their re- 36 stricted scope and "sufficient factual context."' Another exception was Justice Kennedy's discussion of Davis v. Beason.137 He criticized the dissent's reliance on Davis for the proposition that Amendment 2 was "within our constitutional tradition." 138 In Davis, an 1890 decision, the Court observed no constitutional infirmity in a territorial statute that denied voting and office holding rights to, inter alia, persons practicing or advocating polygamy. 139 According to Justice Kennedy, Davis did not provide authority upholding Amendment 2.140 He aeclared that the 1969 case of Brandenburg v. Ohio14 1 invalidated Davis' holding "that persons advocating a certain practice may be denied the right to vote."' 42 Justice Kennedy also stated that Davis' holding, allowing deprivation of voting rights based on status, would require and likely fail strict scrutiny analysis. 143 Finally, Kennedy cited United States Department of Agriculture v. Moreno144 for the proposition that "the desire to harm a politically unpopular group cannot constitute a legitimate governmental interest."'145

D. The Dissent Where the majority saw animus, the dissent saw tolerance, a polity sensitive to the interests of homosexuals. 146 This was apparently because Coloradans no longer send homosexuals to jail for sodomy. 147 What the majority saw as an analysis of a classifi-

136. Romer, 116 S. Ct. at 1627. 137. 133 U.S. 333, 347 (1890), overruled by Romer v. Evans, 116 S. Ct. 1620 (1996). 138. Romer, 116 S. Ct. at 1628. 139. Davis, 133 U.S. at 346-47. 140. Romer, 116 S. Ct. at 1628 ("Davis... is not evidence that Amendment 2 is within constitutional tradition."). 141. 395 U.S. 444 (1969) (per curiam) (reversing the criminal conviction of a Ku Klux Klan leader under a criminal syndicalism statute as violative of the First and Fourteenth Amendments). 142. Ronier, 116 S. Ct. at 1628. However, Justice Kennedy explicitly addressed Davis, emphasizing that Davis' holding involving the denial of rights based on criminal convictions was left undisturbed by Romer, and that this aspect of the Davis holding was irrelevant to Amendment 2. Id. 143. See id. at 1628. 144. 413 U.S. 528 (1973) (striking down a federal regulation denying benefits to unrelated members of food stamp collecting households). 145. Romer, 116 S. Ct. at 1628 (quoting Moreno, 413 U.S. at 534) (emphasis omitted). 146. Id. at 1629 (Scalia, J., dissenting). 147. See id. at 1633 (Scalia, J., dissenting) (citing the fact that Colorado had been one of the first states to repeal its sodomy laws as proof of Colorado's toler- 1997] ROMER V. EVANS cation and the end toward which it is applied, 148 the dissent saw as an argument over policy and morality, one having little to do with status. 149 What the majority described as "sweeping and compre- 151 hensive," 150 the dissent described as "modest." Justice Scalia authored the dissent, and was joined by Chief Justice Rehnquist and Justice Thomas. 5 2 Like the majority, Scalia used the "special rights" issue as the starting point for considering the constitutionality of Amendment 2.153 Unlike the majority, he paid no attention to the Amendment's effects on the coverage of gays, lesbians and bisexuals under civil rights legislation 5 4 Scalia even argued that Amendment 2 did not disfavor Coloradans engaging in homosexual sex. 155 In support, he offered his own characterizations of both the amendment's consequences 156 and the civil rights laws. 157 In considering the consequences of Amendment 2, Justice Scalia contended that homosexuals, like everyone else, continued to enjoy the protections of laws of general applicability, including laws against arbitrary discrimination. 5 8 Drawing support from the Colorado Supreme Court, 159 Scalia maintained that "general laws and policies that prohibit arbitrary discrimination would continue to prohibit discrimination on the basis of homosexual conduct as well."'160 Asserting that the majority "does not dispute all ance). 148. See id. at 1627 (applying this "conventional inquiry"). 149. See id. at 1633-34 (Scalia, J., dissenting) (arguing that Amendment 2 does not prohibit favored status for homosexual individuals, but rather it prohibits favored status for homosexuality). 150. Id. at 1625. 151. Id. at 1629 (Scalia, J., dissenting). Minimizations of this order run throughout the dissent. At the outset, Justice Scalia described Amendment 2 not as aimed at harming homosexuals but as "rather a modest attempt by seemingly tolerant Coloradans to preserve traditional sexual mores against the efforts of a politically powerful minority to revise those mores through use of the laws." Id. at 1629. Later on, he characterized the action taken in Amendment 2 as "the small- est conceivable. The Court's portrayal of Coloradans as a society fallen victim to pointless, hate-filled 'gay bashing' is so false as to be comical." Id. at 1633. 152. Id. at 1629 (Scalia, J., dissenting). 153. See id. at 1629-31. 154. See id. 155. Id. at 1637. 156. See id. at 1630; see also infra text accompanying notes 158, 160-62. 157. See Romer, 116 S. Ct. at 1637; see also infra text accompanying note 163. 158. Romer, 116 S. Ct. at 1630 (Scalia, J., dissenting). 159. Id. (pointing to the Colorado Supreme Court's determination that Amend- ment 2 would have no effect on laws protecting against discrimination based on such factors as age, marital or family status, veterans status and for any legal, off- duty conduct, such as smoking tobacco). 160. Id. Scalia notes, for example, that Amendment 2 would not affect a state Law and Inequality [Vol. 15:275 this, but assumes it to be true,"161 Scalia concluded that the majority's primary thesis was the argument "that any group is denied equal protection when, to obtain advantage (or, presumably, to avoid disadvantage), it must have recourse to a more general and hence more difficult level of political decisionmaking than oth- 2 ers."16 In this discussion, Scalia made no mention of the impact of Amendment 2 on coverage for homosexuals under civil rights laws. This omission implicitly asserts that the protections of civil rights laws are "special." That is, those not covered by civil rights laws, i.e., most individuals, enjoy no protection from the arbitrary actions of those upon whom they depend for employment, housing and access to public accommodations. Scalia objected to the hiring policy of the Association of American Law Schools, for example, because it would allow employment to be denied "because the applicant is a Republican; because he is an adulterer; because he went to the wrong prep school or belongs to the wrong country club, because he eats snails; [or] because is a womanizer," but not because the applicant is homosexual. 163 By distinguishing the last rationale, Scalia insisted, civil rights laws confer "special" status. Justice Scalia also took issue with the majority's argument that requiring homosexuals but no one else to use the more difficult process of constitutional amendment to obtain coverage under civil rights laws involved a denial of equal protection. 164 According to Scalia, "the world has never heard of such a principle, which is why the Court's opinion is so long on emotive utterance and so short on relevant legal citation."' 65 "Requiring that laws singling out homosexuals be henceforth made by constitutional amendment rather than by local or state officials no more violates the Equal Protection Clause than would moving the power to make drug policy or smoking policy from the locality to the state government."'166 The Federal Constitution in fact dictates that some law requirement "that pensions be paid to all retiring state employees with a certain length of service." What Amendment 2 prohibited, he asserted, was policies that would provide a benefit to homosexuals that others did not receive, such as "death benefit payments to the 'life partner' of a homosexual when it does not make such payments to the long-time roommate of a nonhomosexual employee." Id. 161. Id. 162. Id. 163. Id. at 1637. 164. Id. at 1630-31. 165. Id. at 1630. 166. Id. at 1634. According to Scalia, the majority's position "is proved false every time a state law prohibiting or disfavoring certain conduct is passed, because such a law prevents the adversely affected group-whether drug addicts, or smok- 1997] ROMER V. EVANS

groups such as theists will find it more difficult than others to obtain favorable policy.167 Justice Scalia's position depended upon a blurring of the line between a characteristic that attaches to a person and a characteristic that attaches to a person's conduct or circumstance. His view that the class (of gay men, lesbians and bisexuals) was defined by behavior emerged most clearly a few pages later where he argued that Amendment 2 did not "prohibit giving favored status to people who are homosexuals; they can be favored for many reasons-for example, because they are senior citizens or members of racial minorities. But it prohibits giving them favored status because of their homosexual conduct-that is, it prohibits favored 68 status for homosexuality."1 The defining of a group by its behavior bears also on the question of the applicable standard of review and the analysis of rationality. 169 If a state could criminalize gay sex, as the Supreme Court said it could in Bowers v. Hardwick,170 a case Justice Scalia accused the Court of ignoring, 171 then surely it could level less onerous burdens on homosexuals. But even if Hardwick did not fore- close heightened scrutiny, Scalia claimed that gays and lesbians were not a politically powerless group that merited Court protection. 172 If rationality is the standard, the dissent argued, then Amendment 2 should be constitutional. 173 Under the rational basis test it does not matter if the classification is imperfect, and it would therefore be permissible even if it included some who did not engage in the discouraged conduct but had only the inclination to do so. 174 Scalia also took up the governance argument, calling ' 7 5 Colorado's process the "most democratic of procedures."'

ers, or gun owners, or motorcyclists-from changing the policy thus established in 'each of [the] parts' of the State." Id 167. The First Amendment's Establishment Clause, applied against state action through the Fourteenth Amendment, prevents those who would wish to establish a state church from doing so without a constitutional amendment. See U.S. CONST. amend. I. 168. Id. at 1633 (emphasis added). 169. See discussion supra Part II.A. 170. 478 U.S. 186 (1986). 171. Romer, 116 S. Ct. at 1631, 1632 (Scalia, J., dissenting). 172. Id. at 1637. 173. Id. at 1632. 174. See id. 175. Id. at 1634. Law and Inequality [Vol. 15:275

III. Legal and Political Issues In Romer, the Court confronted for the first time an Equal Protection challenge to a classification based on sexual orientation. The issue of homosexuality had been before the Court once before in Hardwick,176 but the plaintiffs in that case had leveled a challenge based on the Due Process Clause. 177 If little else, Romer sig- naled that there were limits to the disadvantages that could be placed on homosexuals. But the decision provided little guidance beyond that. The Court considered but did not adopt the state supreme court's position that Amendment 2 affected the fundamental right of gays and lesbians to participate in the political process, and therefore, required strict scrutiny. 78 It did, however, agree that Amendment 2 made it more difficult for gays and lesbians to have an effective voice in the political process, and that this in itself was "a denial of equal protection of the laws in the most literal 79 sense."1 The Court also did not elevate sexual orientation to a suspect or semi-suspect status, as a number of lawyers and commentators had urged it to do.180 Justice Kennedy did not link the decision in Romer to the growing number of cases that purport to apply the rationality standard, but instead conducted a review that was clearly more searching than the traditional test required.18 ' Jus- tice Kennedy took great care to confine the decision to its facts.

176. Bowers v. Hardwick, 478 U.S. 186 (1986). 177. Hardwick involved a challenge to Georgia's sodomy statute by Michael Hardwick who had been charged, though not ultimately prosecuted, with violating the law. Id. at 187-88. Hardwick asserted that the statute violated the right of privacy protected under the Due Process Clause of the Fourteenth Amendment insofar as it criminalized consensual sodomy. Id. at 188. Hardwick did not raise a claim under the Equal Protection Clause. Id. at 198 n.8. 178. See Romer, 116 S. Ct. at 1624. 179. Id. at 1628. 180. This argument appeared in both the briefs on behalf of gays and lesbians and in testimony in the trials challenging the Colorado and Cincinnati initiatives. See e.g, Equality Found. of Greater Cincinnati, Inc., v. City of Cincinnati, 860 F. Supp. 417, 424-26 (S.D. Ohio 1994) (challenging a citizen-initiated amendment to the City Charter barring laws protecting gays and lesbians from discrimination), affd in part, vacated in part, 54 F.3d 261 (6th. Cir. 1995), vacated, 116 S. Ct. 2519 (1996). Many commentators see homosexuality as a suspect classification, or at least a semi-suspect classification. See, e.g., ELY, supra note 45, at 162-64; Bruce A. Ackerman, Beyond Carolene Products, 98 HARV. L. REV. 713, 742 (1985), Constitu- tional Limits on Anti-Gay-Rights Initiatives, 106 HARv. L. REV. 1905, 1906-07 (1993); John F. Niblock, Anti-Gay Initiatives:A Call for Heightened Judicial Scru- tiny, 41 UCLA L. REV. 153, 188-97 (1993). 181. See discussion infra Part III.A. 1997] ROMER V. EVANS

The Court purposely described its analysis as "exceptional," mentioning twice that Amendment 2 defied traditional analysis and was instead outside both heightened scrutiny and traditional rational basis analysis. 8 2 Because of these disclaimers, it is clear that the Court did not settle on an approach to classifications based merely on sexual orientation in Romner. The opinions in this decision were remarkable in several respects. What is striking about the majority's opinion is its con- scious effort to divorce itself from existing equal protection doctrine. As a consequence, its action appears to be much broader and far less grounded in the law than it actually is, and therefore much more vulnerable to attack. What is also striking about the majority opinion is its apparent indifference to the politics surrounding the case. With the exception of the issue of "special rights," Justice Kennedy failed to confront the dissent's arguments. This will give the dissent an advantage in the dialogue sure to follow in the lower courts, legislatures and public opinion. The dissenting opinion was also unusual because of the extent to which it accepted and expanded upon the political arguments made by the anti-gay forces in the campaign for Amendment 2.

A. Equal Protectionand the Enhanced Rational Basis Test Justice Kennedy treated Amendment 2 as if it could not be analyzed within existing equal protection doctrine. Had the choices been only to apply either strict scrutiny or highly deferential review, that would have been true. The analysis that Justice Kennedy chose fit neither of these models. His analysis was less rigorous than strict scrutiny, involving no requirement of compelling interests or narrowly tailored means; yet it was more searching than the traditional rational basis test. Had Colorado only been required to show that it had some conceivable basis for its ac- 183 tion, the Amendment almost assuredly would have survived. A majority of justices still do not acknowledge that they are 18 4 applying "something more than minimum rationality review."'

182. Romer, 116 S. Ct. at 1627-28. 183. For example, Colorado's interest in advancing religious liberty was at least tangentially connected to the law. The trial court rejected this governmental interest as a defense because it was insufficiently tailored to the ends; however, no such precision is needed under the rational basis test. See, e.g., Railway Express Agency v. New York, 336 U.S. 106, 109-10 (1949) (commenting that it is not the function of the Court to "pass judgment on the wisdom [of the ordinance in question]"). See also supra notes 58-59 and accompanying text. 184. City of Cleburne v. Cleburne Living Ctr., Inc., 473 U.S. 432, 459 (1985) (Marshall, J., concurring) (noting that the Court had previously upheld similar Law and Inequality (Vol. 15:275

In practice, however, rationality review now seems to represent more than one standard. It looks increasingly like the sliding scale proposed by Justice Thurgood Marshall in Dandridge v. Wil- liams:l8 5 The less important the interest, the less likely a classification is based upon prejudice, the more likely it is connected with some legitimate state objective and the more deferential the review. 186 The more important the interest, the higher the "degree to which it is infringed,"'187 the more likely the classification is based on prejudice and the stricter the review. 188 Some classifications seem more problematic than others. These include classifications involving a group that "has been subjected to a 'tradition of disfavor" 189 and classifications that, although sometimes relevant to a valid public purpose, most often are not.190 However, while the Court in deciding Romer did not conform to the most deferential type of rational basis analysis, the approach it took was not without precedent. As noted, the Court has increasingly been unwilling to abdicate totally a review which is outside the areas of suspect or semi-suspect classifications or of fundamental rights.'91 This "move away from near-absolute deference to legislative judgment ' 192 has been particularly evident over the past decade. The Court began moving away from its traditional deference in United States Department of Agriculture v- Moreno,193 in which, purporting to apply the rational basis test, the Court struck down a regulation which prevented unrelated individuals living together from collecting food stamps. 194 A year later in Jimenez v. Weinberger,195 the Court invalidated a provision of the Social Security Act which made some but not all illegitimate 196 children ineligible for a parent's disability benefits. Since that time, several other cases have found violations of the Equal Protection Clause while purporting to use the rational imprecision in ordinances under mere rationality review). 185. 397 U.S. 471, 521 (1970) (Marshall, J., dissenting) (upholding a Maryland law capping AFDC payments regardless of the size of the family). 186. See id. at 521-22. 187. TRIBE, supra note 50, § 16-33, at 1610. 188. City of Cleburne, 473 U.S. at 460-61 (Marshall, J., concurring). 189. Id. at 453 (Stevens, J., concurring). 190. Id. at 454. 191. See supra notes 184-90 and accompanying text. See also discussion supra Part II.A. 192. TRIBE, supra note 50, § 16-3, at 1445. 193. 413 U.S. 528 (1973). 194. Id. 195. 417 U.S. 628 (1974). 196. Id. 19971 ROMER V. EVANS basis test. 197 In Hooper v. Bernalillo County Assessor,198 the Court invalidated a New Mexico statute which treated Vietnam veterans who were New Mexico residents prior to 1976 differently from other Vietnam veterans by allowing them a $2,000 annual tax exemption not available to the later residents in the state. 199 Alaska's effort to favor long-term over short-term residents met an identical fate for similar reasons in Zobel v. Williams.200 Texas' law barring the children of illegal aliens from receiving a free public education was struck down in Plyer v. Doe.201 In City of Cle- burne v. Cleburne Living Center, Inc., 202 the Court found that the city acted unconstitutionally when it denied a special use permit 20 3 for the operation of a group home for the mentally retarded. However, most classifications challenged under the Equal Protec- tion Clause still were reviewed under the more deferential standard.20 4 Despite this tendency toward deference, the Court would sometimes apply de facto heightened scrutiny, giving the Romer decision a foundation in the law. A strong argument can be made that Amendment 2 was just the sort of measure for which the heightened scrutiny of the type described by Justice Marshall was appropriate. 2 5 Gays, lesbians

197. Cf. discussion supra Part II.A (indicating that the traditional and most commonly used rational basis test means essentially no review at all). 198. 472 U.S. 612 (1985). 199. Id. 200. 457 U.S. 55, 56-58 (1982) (invalidating an Alaska law which favored long- time residents in the distribution of income from natural resources) See also Alle- gheny Pittsburgh Coal Co. v. Webster County, 488 U.S. 336 (1989) (invalidating a system of assessing property tax which discriminated against recent home pur- chasers); Metropolitan Life Ins. Co. v. Ward, 470 U.S. 869 (1985) (holding that a state cannot discriminate between out-of-state and instate insurance companies in imposing a gross premiums tax rate); Williams v. Vermont, 472 U.S. 14 (1985) (striking down a Vermont automobile use tax that favored residents over non-residents). 201. Plyer v. Doe, 457 U.S. 202 (1982) (employing a rationality test after finding that education was an important but not fundamental interest and that illegal aliens did not constitute a suspect class). 202. 473 U.S. 432 (1985). 203. Id. at 447, 450. 204. See, e.g., Vance v. Bradley, 440 U.S. 93 (1979) (upholding a mandatory retirement age that applied to Foreign Service Officers but not to Civil Service employees, including those stationed abroad); United States R.R. Retirement Bd. v. Fritz, 449 U.S. 166 (1980) (upholding a law that gave a windfall in retirement benefits to some workers whose service was less than others but denied the windfall depending on whether they had worked for the railroad in 1974). 205. See Dandridge v. Williams, 397 U.S. 471, 521 (1970) (Marshall, J., dissenting) (advocating inquiry into the "facts and circumstances behind the law"). See also supra notes 185-87 and accompanying text. Law and Inequality [Vol. 15:275 and bisexuals have clearly been traditionally disfavored groups.20 6 For most of the country's history, homosexuality and homosexual sex have been seen as deviant and immoral.20 7 The medical community long viewed homosexuality as a disease. 208 Even now, those whose homosexual orientation becomes known often face 20 9 discrimination in employment, housing, school and religion. Le- gal discrimination against gays, lesbians and bisexuals has been extensive. 210 Although it was argued in Romer that regardless of the historical discrimination, gays and lesbians no longer face discrimination, and that they have indeed become privileged groups, 211 this does not reflect reality. Sexual orientation is often still a permissible basis for discrimination both in the public and private spheres, and many gays and lesbians still experience discrimination and even violence. 212 Sexual orientation is often irrelevant to legitimate policy aims having little to do with the individual's ability to function in a job or part of civic life. With respect to Amendment 2, the degree of harm inflicted on homosexuals was severe and far-reaching, as noted by the majority opinion.213 It therefore was not necessary (Justice Scalia's dissent notwithstanding) that homosexuals be politically powerless or

206. This is the argument made by those who have favored treating sexual orientation as a suspect classification. See supra notes 61-68. Cass Sunstein agrees that heightened scrutiny is appropriate for some of the same reasons but has argued against suspect classification analysis because of the belief that moving too much beyond public opinion would generate a backlash, which in the end would result in fewer rights for homosexuals. See Cass R. Sunstein, Homosexuality and the Constitution, 70 IND. L. REV. 1, 7-9 (1994). 207. See Stephen Zamansky, Colorado's Amendment 2 and Homosexuals' Right to Equal Protectionof the Law, 35 B.C. L. REV. 221, 244, n.248 (1993). 208. Id. 209. Id. at 245 n.252 (citing Civil Rights Act Amendments of 1981: Hearings on H.R. 1454 Before the Subcoinm. on Employment Opportunities of the House Conmn. on Educ. and Labor, 97th Cong., 2d Sess. 23 (1982) (statement of Rep. Paul Tson- gas of Massachusetts). 210. Same-sex intimate relationships are not recognized by law. See Niblock, supra note 180, at 159. That in turn has economic and social consequences. Same-sex domestic partners are not entitled to health insurance benefits, to share pensions, to inherit absent direct instructions in a will, to visit their partner when ill and so on. Id. at 160. Adoptions by gay and lesbian couples are disfavored. Id. Homosexuals are often disfavored as well in custody disputes. Id. Homosexuals cannot openly serve in the Armed Forces. Id. at 161. 211. See Romer v. Evans, 116 S. Ct. 1620, 1637 (Scalia, J., dissenting). 212. For an extensive examination of sexual orientation discrimination in employment, see Mison et al., supra note 23, at 33. See also Zamansky, supra note 207, at 244-49 (describing status issues still confronting homosexual men and women in society); Garfield, supra note 24, at 730-32 (arguing that homosexuals continue to face difficulties in participating in the political process). 213. Romer, 116 S. Ct. at 1624-27. 1997] ROMER V. EVANS have low incomes in order to qualify for enhanced rational basis review. 214 Ample factors were present to support such heightened 215 review. Although Justice Kennedy did not invoke precedent as a prominent feature of his analysis in Romer, he could have used a number of the cases already mentioned to strengthen his arguments. For example, Romer was not the first case to indicate that antipathy toward the group that is harmed is an unacceptable basis for government action. When the Court invalidated the City of Cleburne's denial of a special use permit for a group home for the mentally retarded, it did so because it believed that the city's decision "rest[ed] on an irrational prejudice against the mentally retarded."216 Romer bears a marked resemblance to City of Cle- burne, Moreno and Plyler, cases in which there was concern that 17 prejudice operated in the creation of the classification. 2 In recognizing that there is a difference between disadvantaging a group in the pursuit of a larger goal and disadvantaging the group for the sole purpose of treating them differently, the Court also had ample support in prior commentary regarding the Equal Protection Clause.218 The interpretation of this clause has been based on a principle of fairness: the idea that government should treat individuals equally unless there is a legitimate and

214. See Dandridge v. Williams, 397 U.S. 471, 521 (1969) (Marshall, J., dissenting) (suggesting that the severity and breadth of a regulation are relevant factors triggering more searching inquiry). 215. As District Judge Spiegel wrote, "[R]elative political power cannot even be a particularly weighty factor, let alone a controlling one. For example, it cannot be said that males, as a group, have been relegated to such a position of political powerlessness as to require special judicial protection." Equality Found. for Greater Cincinnati, Inc. v. City of Cincinnati, 860 F. Supp. 417 (S.D. Ohio 1994). Justice Powell's comments in San Antonio Independent School Board v. Rodriguez, 411 U.S. 1, 28 (1973), indicate that he considered political powerlessness to be but one of the factors triggering heightened scrutiny. Explaining why heightened scrutiny was not necessary in Rodriguez, he noted that the inhabitants of the poor districts were saddled with "none of the traditional indicia of suspectness: the class is not saddled with such disabilities, or subjected to such a history of purposeful unequal treatment, or relegated to such a position of political powerlessness as to command extraordinary protection from the majoritarian political process." Id. at 28 (emphasis added). But see Ronald Dworkin, Sex, Death, and the Courts, N.Y. REV. OF BOOKS. Aug. 8, 1996, at 44 (asserting that access to the political process is not a fundamental right). 216. City of Cleburne v. Cleburne Living Ctr., Inc., 473 U.S. 432, 450 (1985). 217. See discussion supra notes 184-97 and accompanying text. 218. See e.g., ELY, supra note 45, at 159-70 (explaining the characteristics that make a classification worthy of more searching inquiry under the Equal Protection Clause); Kenneth L. Karst, Suspect Classifications, in CIVIL RIGHTS AND EQUALITY 285, 285-87 (Leonard W. Levy et al. eds., 1989) (giving a brief overview of classes treated as "suspect" by the Court). 308 Law and Inequality [Vol. 15:275 non-prejudicial reason to do otherwise. 219 The concept of "suspect" classifications arose in part out of the understanding that some 220 classifications carry a high risk of prejudicial decisionmaking. As Justice White wrote in City of Cleburne, the general deference accorded legislative choice of classifications "gives way... when a statute classifies by race, alienage, or national origin. 221 Justice White reasoned that such classifications "are seldom relevant to the achievement of any legitimate state interest that laws grounded in such considerations are deemed to reflect prejudice and antipathy of a view that those in the burdened class are not as 222 worthy or deserving as others." The highly deferential rational basis review which controls most classifications was based upon concerns about governability, federalism and democracy.2 23 But even the "theory of rationality as governing the relation between means and ends assumes that all legislation must have a legitimate public purpose or set of purposes based on some conception of the general good."2 24 Therefore, even under rational review, "the legislature is not entitled to pick out a group it disfavors, declare that group to be different and then impose a special ... burden on the disfavored group."2 25 Or in the words of Justice Stevens, "the word 'rational'-for me at least- includes elements of legitimacy and neutrality that must always characterize the performance of the sovereign's duty to govern im- 6 partially." 2 It is not clear why the Court chose to ignore these cases and commentary.22 7 The Court may have felt that treating Romer as

219. Karst, supra note 218. See also Kenneth L. Karst, Forward:Equal Citizen- ship Under the FourteenthAmendment, 91 HARV. L. REV. 1 (1977). 220. TRIBE, supra note 50, § 16-2, at 1445. 221. City of Cleburne, 473 U.S. at 440. 222. Id. 223. TRIBE, supra note 50, § 16-2, at 1446-50 (mentioning considerations of governance and democracy); San Antonio Indep. Sch. Bd. v. Rodriguez, 411 U.S., 44 (1973) (explaining the need to recognize the effect that claims arising under the Equal Protection Clause have on the relationship between national and state power under our federal system). 224. TRIBE, supra note 50, § 16-2, at 1440. 225. ROTUNDA & NOWAK, supra note 49, § 18-3, at 36. 226. City of Cleburne, 473 U.S. at 452 (Stevens, J., concurring). 227. The origin of the Court's position that Amendment 2 created a situation that was unprecedented is not hard to locate. It looked much like the argument that Laurence Tribe had advanced in an amicus brief. See Brief of Laurence H. Tribe et al., as amici curiae in support of Respondents at 1-3, Romer v. Evans, 116 S. Ct. 1620 (1996) (No. 94-1039). Locating the argument's source, however, does little to explain why it was adopted or why other clearly significant cases were ig- nored or simply not discussed. 1997] ROMER V. EVANS exceptional confined its reach. No doubt the Justices saw the danger in taking a step that was too bold in this highly controversial area. However, treating the decision as if it had little relationship to precedent exposed the Court to just the type of attack that 2 28 Scalia leveled in his dissent.

B. Homosexuals as an Independently Identifiable Group The Court's opinion in Romer assumes as a premise that a law singling out individuals on the basis of their sexual orientation identifies them by their personal status. From this perspective the debate over the law was whether there had been discrimination, and if so, whether the discrimination was permissible. The dissent challenged this framework. Justice Scalia essentially defined homosexuality as conduct rather than status, and thus depicted Amendment 2 as aimed at conduct rather than personal identity. 229 Such a formulation profoundly alters the argument and the outcome. If Amendment 2 merely targeted behavior, the Equal Protection Clause would not apply, and the Constitution would be silent unless the targeted activity was protected under the Due Process Clause of the Fourteenth Amendment. 23 0 In addition, there could be no claim based on limiting access to the political process. Justice Scalia, however, was not correct. There is an appropriate line to be drawn between a classification based on status and a classification based on behavior. The difference can best be seen by reference to characteristics about which there would be no dispute, such as race and gender. These characteristics define who we are. They are personal traits that transcend any position on a particular issue, and which do not change. At the other end of the

228. See supra Part II.D (discussing Justice Scalia's criticisms and arguments). 229. This offered apparent advantages for Justice Scalia because merging status and conduct casts a more favorable light on classifications that single out homosexuals for disfavorable treatment. Cf. Holmes v. California Army Nat'l Guard, 920 F. Supp. 1510 (N.D. Cal. 1996). The district court, ordering the reinstatement into the California National Guard of First Lt. Andrew Holmes, saw the "don't ask, don't tell" policy in a similar vein. Id. at 1534. The court in its review of that case observed that the drafters of the "don't ask, don't tell" policy "merged the concepts of status and conduct in an attempt to circumvent the constitutional proscrip- tions." Joining conduct and status has been used elsewhere to defend such policies. See, e.g., Cammermeyer v. Perry, 97 F.3d 1235, 1237 (9th Cir. 1996) (discussing the relationship between homosexual status and homosexual sexual conduct). 230. This potential exposure is most apparent in Bowers v. Hardwick, 478 U.S. 186 (1986), which determined that the conduct of homosexual sodomy was not constitutionally protected. Law and Inequality [Vol. 15:275 spectrum would be legislative categories that distinguish between types of economic activities. Whether or not a person is allowed to organize a business in a particular manner certainly affects that person, but it does so without any attention to the individual's personal attributes. It is inaccurate to say that any attribute which involves behavior is merely a personal choice of conduct, rather than a personal trait. Some characteristics tied to personal identity can have a behavioral component. Perhaps the clearest example of such a trait is religion. It often affects behavior but it neither totally pre- dicts it nor is confined to it. Religion involves choice to some extent, but for most people it is a choice that is determined by the religious affiliation of their family. There is a sense of permanence to it. This trait is semi-permanent, only partially a matter of choice, potentially disconnected from behavior, and descriptive more of who the person is than what the person does. Even when legislation aims at behavior, which arose from choice and may be 231 amenable to change, it may still involve a regulation of status. Sexual orientation is likewise a personal attribute that goes beyond conduct. One need not hold that sexual orientation is immutable or biologically determined to see it as a personal attribute. Moreover, there is increasing evidence that supports the view that sexual orientation often involves a genetic component.232 Like religion, it is often a central aspect of an individual's identity and it is more or less permanent. 233 Sexual orientation is separate from

231. See Robinson v. California, 370 U.S. 660, 666 (1962) (holding unconstitutional a law that criminalized drug addiction, a status). Justice Powell's clerks apparently tried to use the case to convince him to support Hardwick. See JOHN C. JEFFRIES, JR., JUSTICE LEWIS F. POWELL, JR. 520 (1994). According to Jeffries, had Georgia "punished the status of being homosexual," the ruling would have been applicable. "[But] that the State had not tried to do." John C. Jeffries, Changing Times: Gay Rights, in HOMOSEXUALITY 103, 110 (Robert M. Baird & M. Katherine Baird eds., 1995) [hereinafter Changing Times]. 232. The biological determinants of sexual orientation are not yet well understood, but as discussed earlier, there is a growing body of evidence that, for many, homosexual orientation is a trait which becomes manifest early in life, is not freely chosen, and is not amenable to change. See Zamansky, supra note 207, at 241-44 (surveying scientific evidence that homosexuality is immutable). 233. Many commentators have argued that sexual orientation should be treated as an immutable characteristic. See id. For Samuel Marcosson, what matters is not whether sexual orientation is immutable for all, but that it is so for many: Admittedly, sexual orientation may not be immutable for all gay men and lesbians. The point is that the testimony of countless people regarding their perceptions of their sexual orientations, together with emerging evidence of a link between biology and sexual orientation, makes it clear that at least for many gays, sexual orientation is immutable, an essential aspect of self. 1997] ROMER V. EVANS particular policy positions. It often bears no relationship to the ability to participate in the general activities of social and economic life. Such a trait suffices to define an "independently identifiable group," i.e., a group whose characteristics in some real sense transcend "the mere support for a single political issue, such as 234 race, gender or sexual orientation." Amendment 2 explicitly includes within its scope not just people engaging in homosexual sex, but also those of homosexual, lesbian or bisexual orientation.235 Anti-gay forces have argued that this is a distinction without a difference. 236 Justice Scalia en- dorsed this view in his dissent.237 However, there are significant periods of time in people's lives when they do not engage in any sexual activity regardless of their orientation. The dissent glossed over this difficulty by fusing the discussion of Amendment 2 as a regulation of conduct with a discussion of the rationality of singling out gays, lesbians and bisexuals for disfavorable treatment even if they have not engaged in homosexual sex.238 This approach was improper because Amendment 2 does not directly address homosexual orientation as it affects the workplace or any other area of potential discrimination. Instead, it aims at an entire class of people having an attribute arguably beyond their free choice, ir- respective of their conduct.

Marcosson, supra note 25, at 177; see also The ConstitutionalStatus of Sexual Ori- entation: Homosexuality as a Suspect Classification,98 HARV. L. REV. 1285, 1302 & n.91 (1985) (arguing that uncertainty as to the origin of homosexuality favors treatment as an immutable characteristic). 234. Equality Found. of Greater Cincinnati v. City of Cincinnati, 860 F. Supp. 417, 434 n.12 (S.D. Ohio 1994). 235. COLO. CONST. art. II, § 30b ("Neither the State of Colorado ... nor any of its agencies ... shall enact, adopt or enforce any statute, regulation, ordinance or policy whereby homosexual, lesbian or bisexual orientation, conduct, practices or relationships . . . [shall] entitle any person . . . [to] claim . . . discrimination.") (emphasis added). 236. See, e.g., Romer v. Evans, 116 S. Ct. 1620, 1632 (1996) (Scalia, J., dissenting) C'Some courts of appeals have concluded that, with respect to laws of this sort at least, that is a distinction without a difference.") (citing Equality Found. of Greater Cincinnati v. City of Cincinnati, 54 F.3d 261, 267 (6th Cir. 1995)). 237. Id. at 1632 (Scalia, J., dissenting) (supporting the Supreme Court of Colo- rado's assertion that the personal characteristics of sexual orientation, conduct, practices and relationships are "not truly severable from one another because each provides nothing more than a different way to identify the same class of persons") (quoting Evans v. Romer, 882 P.2d 1335, 1349-50 (Colo. 1994) (emphasis added)). 238. See id. at 1632 (Scalia, J., dissenting) (arguing that because Bowers v. Hardwick, 478 U.S. 186 (1986), established that "it is rational to criminalize the conduct, surely it is rational to deny special favor and protection to those with a self-avowed tendency or desire to engage in the conduct"). Law and Inequality [Vol. 15:275

C. "Special Rights" or "EqualProtection'? The question of whether civil rights laws provide "special rights" was as central to the constitutional controversy as it had been to the political controversy over Amendment 2. In the context of homosexual rights, the "special rights" label was particularly inaccurate. As Justice Kennedy indicated, the reason that states and localities enacted anti-discrimination laws was to counter discriminatory practices. 239 There would be no need for such laws if general prohibitions against arbitrary action were sufficient to put all citizens on an equal footing. The "special rights" language used by anti-gay forces suggests that gays, lesbians and bisexuals were seeking rights that others did not have. In fact, what they were asking for was decision-making in which sexual orientation would not count.240 Put differently, what homosexuals sought when they argued for protection against discrimination was to be treated like everyone else. One argument that supports the notion of "special rights" is that an employer has a right to run his business as he pleases, and that "a person can ordinarily be fired for a wide range of reasons but not for being a member of a protected group."241 As a result, "the government is choosing to favor the equality-based right of one person over the liberty-based right of another-thus conferring a 'special right' for the former."242 This argument is flawed in at least two respects. First, as the Court pointed out in Romer, common law limited the so-called private right to discriminate. 243 Sec- ond, the comparison between employer and employee is one between differently situated people, and thus not relevant to equal protection analysis. The civil rights laws were passed in response to discriminations that the majority of people do not face. 244 Such

239. Id. at 1624-25. 240. See id. at 1626 ('It is a fair, if not necessary, inference from the broad language of the amendment that it deprives gays and lesbians even of the protection of general laws and policies that prohibit arbitrary discrimination in governmental and private settings."). For example, Amendment 2 might be construed to strip homosexuals of such general protections as judicial review of agency action under the "arbitrary and capricious" standard. Id. 241. Schacter, supra note 38, at 304-05. 242. Marcosson, supra note 25, at 145. 243. Romer, 116. S. Ct. at 1625. 244. See, e.g., Sperling v. Hoffman-La Roche, Inc., 24 F.3d 463, 472 n.14 (3d Cir. 1994) (asserting that many civil rights laws exist to ban age discrimination in employment); cf. Casey v. City of Cabool, Mo., 12 F.3d 799, 805 (8th Cir. 1993) (articulating Congress' intent regarding civil rights actions as including the desire to shift the cost of attorneys' fees to guilty parties and thus enable an individual to act as a "private attorney general"). 1997) ROMER V. EVANS

laws merely put groups with recognized disadvantages on an equal footing with other groups. Whatever could be said for the "special rights" view regarding private discrimination loses considerable force when applied to the government. This is because government has no cognizable liberty interest against which the interest in equal treatment competes. Thus Justice Kennedy's observation that Amendment 2 was "[n]ot confined to the private sphere" is important.245 In fact, the Colo- 246 rado Supreme Court itself gave two examples. Amendment 2 mischaracterizes anti-discrimination laws as vehicles for preferences and quotas.247 It purports to right these wrongs by barring laws that take account of any sexual orientation, whether homosexual, bisexual or straight, as a basis for decision.248 The Court might have pointed out that striking this seemingly neutral language had little, if any, connection to quotas or preferences. By ignoring the affirmative action mischaracteri- zation, the Court's opinion failed to avert "the public's hostility toward the most powerful symbol of civil rights backlash,"249 and thus compromised some of its persuasiveness. D. Animus/Moral Disapproval as a Basis for Discrimination The dissent's most compelling argument turns out to be one that was downplayed in the political campaign for Amendment 2: that discrimination is in fact permissible. Justice Scalia claimed first that that the Court's position in striking down Amendment 2 was inconsistent with its willingness to allow States to criminalize homosexual sodomy as it had in Bowers v. Hardwick.250 According to Justice Scalia the removal of homosexuals from protected status was certainly less severe than actually punishing them for their

245. Romer, 116 S. Ct. at 1626. 246. Id. (The State Supreme Court cited two examples of protections in the government sphere that are now rescinded and may not be reintroduced."). The rescinded protections had prohibited discrimination in the areas of state employment and state colleges. Id. (citing Evans v. Romer, 854 P.2d 1270, 1284-85 (Colo. 1993)). 247. See COLO. CONST. art. II, § 30b (mentioning preferences and quotas explicitly despite their tenuous relation to anti-discrimination laws). 248. See Marcosson, supra note 25, at 141-42 (pointing out that such classifications are, in a sense, facially neutral). 249. Schacter, supra note 38, at 304. 250. Romer, 116 S. Ct. at 1629 (Scalia, J., dissenting) (citing Bowers v. Hard- wick, 478 U.S. 186 (1986)). Law and Inequality [Vol. 15:275 sexual conduct, as had been allowed earlier. 251 Under this reasoning, Amendment 2 helped Coloradans express their disapproval of homosexual orientation at a level clearly permitted by the Con- stitution. Bowers v. Hardwick 52 was a due process case, however, and there is some question whether Hardwick would have come out the way it did if an equal protection claim had been made. 253 As one prominent constitutional scholar has written, "[b]ecause of the independence of the lines of analysis under the two clauses, the conclusion in Hardwick does not dispose of the issue. 254 But even if the Court were to conclude that a state may punish homosexual and not heterosexual sodomy, it does not follow that a state may broadly disadvantage anyone who might merely be inclined to commit such sodomy. It would be a violation of due process to punish someone for an act they have not yet committed.25 5 Also, the penalty exacted under Amendment 2 was arguably more severe than those under the criminal laws. Amendment 2 was very broad and burdened gays and lesbians in a vast array of activities. Additionally, to the extent that the Romer Court relied on the right to political participation in reaching its decision, the Hard- wick case had no relevance. Similar objections can be made to Justice Scalia's analogy to polygamy. 25 6 Society can indeed enact laws which punish polyga-

251. Id. at 1631 (Scalia, J., dissenting) ("If it is constitutionally permissible for a State to make homosexual conduct criminal, surely it is constitutionally permissible for a State to enact other laws merely disfavoring homosexual conduct."). 252. 478 U.S. 186 (1986). 253. The Court's rejection of the due process claim included a comment that no equal protection issue had been presented, thus leaving open the possibility that such a claim might be mounted in the future. Id. at 196 n.8. Obviously, the fact that a bare majority (5-4) decided Hardwick enhances this possibility. The majority, moreover, was particularly unstable. See ChangingTimes, supra note 231, at 105-18 (pointing to Justice Powell's ambivalence throughout the deliberations and his later recantation). Had equal protection arguments been presented in Hard- wick, the outcome may well have been different. 254. Cass R. Sunstein, Sexual Orientation and the Constitution: A Note on the Relationship Between Due Process and Equal Protection, 55 U. CHI. L. REV. 1161, 1163 (1988). 255. The Court is currently grappling with whether it is permissible to restrain sex offenders who appear to be a threat to the community. See In re Care and Treatment of Hendricks, 912 P.2d 129 (Kan. 1996), cert. granted, Kansas v. Hen- dricks, 116 S. Ct. 2522 (1996). A decision for the State would probably characterize the restraint as noncriminal and non-penal. It would nevertheless be based on predicted future behavior, though in circumstances where the State's interest was compelling. 256. See Romer, 116 S. Ct. at 1635 (Scalia, J., dissenting) (likening homosexuality to polygamy). 1997) ROMER V. EVANS mists, but such laws involve conduct, not inclination. While most gays and lesbians are likely to engage in sexual acts with persons of the same sex, surely there are some who do not, just as there are some heterosexuals who do not engage in sex. There are in fact many gays and lesbians who are not sexually active at any particular time. The Court did make clear that a law disadvantaging those of polygamous orientation-those who would practice polygamy if they could-would not withstand constitutional scru- 57 tiny.2 Regardless of precedent, there remains the broader assertion that discrimination is permissible for no reason other than the majority's desire to disapprove of homosexuality. 258 This argument concerns the proper reading of the Constitution. One side assumes, as did Justice White in Hardwick, that it is "legitimate for a state to impose a disadvantage on a particular group just to express the majority's moral contempt for that group's practices, even when there is no other proper purpose."259 The other side be- lieves that the state can impose disadvantage only in the face of actual harm from the group's activities. 260 Even if one accepts the former position, though, the argument is not compelling in the context of homosexual rights. As noted previously, Amendment 2 burdened not just homosexual conduct, but homosexual orientation as well.2 6 1 Simple fairness demands that "people should not be ill-treated based on a characteristic or status they are unable to

257. Id. at 1628. 258. The dissent erroneously characterized the Court's position as having "placed the prestige of this institution behind the proposition that opposition to homosexuality is as reprehensible as racial or religious bias." Id. at 1629. Had this characterization been accurate, sexual orientation would have been elevated to a suspect classification. See supra notes 61-63 (discussing the importance of such a classification with respect to the standard of review). 259. Dworkin, supra note 215, at 49 (characterizing Justice White's opinion in Hardwick). 260. See id. (characterizing Justice Kennedy's opinion in Romer). The dissent in Roiner argued that what Justice Kennedy called "animus" was nothing more than a statement of moral disapproval. Id. (citing Romner, 116 S. Ct. at 1633 (Scalia, J., dissenting)). Justice Kennedy did not respond to this part of the dissent, possibly because the majority did not hold to one view. It is not unknown for justices who join an opinion to do so for reasons that are quite different. This often results in an opinion that is not reasoned as tightly as it otherwise might have been. See, e.g., RICHARD KLUGER, SIMPLE JUSTICE: THE HISTORY OF BROWN V. BOARD OF EDUCATION AND BLACK AMERICA'S STRUGGLE FOR EQUALITY 702-05 (1975) (pointing to the terse, bland reasoning expressed in the controversial but unanimous opinion of Brown, 347 U.S. 483 (1954)). 261. See Part III.B supra for arguments that orientation, not conduct, is the correct focus for analyzing homosexual discrimination issues. Law and bIequality [Vol. 15:275

' change. "262 And there is increasing evidence "that at least for 263 many gays, sexual orientation is immutable."

IV. An Alternate Approach-Participation in the Political System as a Fundamental Right Thus far, this Comment has argued that the Court's decision in Romer v. Evans was justified as a matter of law, but that its legal analysis would have been stronger if it had not treated Romer as an anomaly. There was ample precedent available within existing equal protection case law. 264 Such omissions are often of interest only to legal scholars, but this was a highly visible case that will probably have significant repercussions. Weakness in the majority opinion can result in insufficient guidance to lower courts and to other political actors. 265 It also can harm the Court's influence by making the Court appear unprincipled. As noted earlier, the Court actually used an enhanced rational basis review in its Romer decision, although it did not say so explicitly. 266 Although this was defensible, the Court would have had an easier time had it followed the Colorado Supreme Court and declared that there was a fundamental right to equal participation in the political process which was limited by Amendment 2. Because such an approach involves a proven constitutional argument, it would have been stronger politically.

A. Positionof the Colorado Supreme Court The Colorado Supreme Court's analysis of Amendment 2 started from the proposition that "the Equal Protection Clause of the United States Constitution protects the fundamental right to participate equally in the political process. '267 "Any legislation or state constitutional amendment which infringes on this right by 'fencing out' an independently identifiable class of persons must be subject to strict judicial scrutiny."2 68 The court determined that:

262. Marcosson, supra note 25, at 174. 263. Id. at 177 (emphasis omitted). 264. See supra Part III.A (providing several applications of enhanced rational basis review). 265. See TRIBE, supra note 50, § 1-9, at 15-17 (arguing that the fundamental indeterminacy of constitutional interpretation has been a significant contributor to the Court's role as the definitive interpreter of the Constitution). 266. See supra Part III.A. 267. Evans v. Romer, 882 P.2d 1335, 1339 (Colo. 1994) (quoting Evans v. Romer, 854 P.2d 1270, 1282 (Colo. 1993)), affd, Romer v. Evans, 116 S. Ct. 1620 (1996). 268. Id. 1997] ROMER V. EVANS

Amendment 2 alters the political process so that a targeted class is prohibited from obtaining legislative, executive, and judicial protection or redress from discrimination absent the consent of a majority of the electorate through the adoption of a constitutional amendment. Rather than attempting to withdraw anti-discrimination issues as a whole from state and local control, Amendment 2 singles out one form of discrimination and removes its redress from consideration by the 269 normal political processes. Accordingly, the court applied strict judicial scrutiny, not because homosexuals constituted a suspect class, but because the right affected was fundamental. The court then held that Colo- rado had not offered sufficient justification to sustain the amendment.270 The U.S. Supreme Court did not rely heavily on this fundamental rights analysis; however, Justice Kennedy's reference to the discriminatory nature of Amendment 2 and the obstacles it imposed on the ability of gays, lesbians and bisexuals to influence the political process reflects at least a limited recognition that the state Supreme Court's position had some merit.27 1 Not supris- 272 ingly, Justice Scalia rejected this approach. Despite commentary to the contrary, the state court's position did not involve a novel proposition. 273 The U.S. Supreme Court consistently has been concerned when the State restructures its government in a way that makes it more difficult for an "independently identifiable group"274 to have influence over the machinery of government. The Court's effort to assure a fair po- 275 litical process can be seen in a number of its voting rights cases.

269. Id. 270. For a discussion of the Colorado Supreme Court's position see supra Part II.B. 271. Ronier, 116 S. Ct. at 1629. 272. Id. at 1634-35. 273. Ronald Dworkin's analysis necessarily accepts the argument that there is no difference between identifiable groups and independently identifiable groups. See Dworkin, supra note 215, at 44 (arguing against the Colorado Supreme Court's holding that Amendment 2 implicated a fundamental right). 274. Gordon v. Lance, 403 U.S. 1, 5 (1971) (using this terminology to signify particular minority groups-racial, religious or otherwise-subjected to Virginia's discriminatory voting policy). 275. See Davis v. Bandemer, 478 U.S. 109 (1986) (indicating that political ger- rymanders might contravene equal protection principles); Lubin v. Panish, 415 U.S. 709 (1974) (invalidating a candidate filing fee as applied to indigents); Kramer v. Union Free Sch. Dist., 395 U.S. 621 (1969) (striking down a state plan that limited voting in a school district to property owners or people with children in the schools); Williams v. Rhodes, 393 U.S. 23 (1968) (striking down a requirement that new parties file petitions nine months prior to the election); Reynolds v. Simms, 377 U.S. 533 (1964) (invalidating an apportionment scheme that failed to provide for one person/one vote, noting that vote dilution debased democracy). Law and bequality [Vol. 15:275

The Fourteenth Amendment has been interpreted to protect against not only denials of the vote but also dilution of the vote. In Goinillion v. Lightfoot,276 for example, the Court overturned Ala- bama's efforts to redraw voting districts based on its understanding that the purpose was to dilute the political power of African Americans. The argument that these protections are confined to suspect classifications is inaccurate. It is true that cases in which the Court has invalidated initiatives have involved racial classifications;277 however, the Court's language has suggested a broader reading. Consider the Court's observation that "the State may no more disadvantage any particular group by making it more difficult to enact legislation in its behalf than it may dilute any person's vote or give any group a smaller representation than another of comparable size."278 The Court has also used the phrase "independently identifiable groups," 279 which could certainly be 2 0 applied to homosexuals. Finally, although James v. Valtierra 8 is generally used to support the notion that only racial classifications are forbidden, James is distinguishable from Romer because it did 28 not deal with an independently identifiable group. ' The principle that the State should not make it more difficult for some citizens than others to influence the political process is

276. 364 U.S. 339 (1960) (invalidating efforts to redraw the voting districts in and around Tuskeegee, Alabama, which were intended to dilute the voting power of African Americans). 277. See, e.g., Washington v. Seattle Sch. Dist. No. 1, 458 U.S. 457 (1982) (invalidating a state initiative that shifted power to remedy de facto racial segregation from the City of Seattle to the State of Washington) and Hunter v. Erick- son, 393 U.S. 385 (1969) (striking down a city charter amendment adopted by a citizen initiative which made the adoption of housing anti-discrimination legislation more difficult). 278. Hunter, 393 U.S. at 393. 279. See Gordon v. Lance, 403 U.S. 1, 5, 7 (1971) (linking the violation of the Equal Protection Clause to a voting provision's propensity to "fence ouf' a sector of the population). 280. 402 U.S. 137 (1971) (upholding the constitutionality of making it more difficult for low-income individuals to obtain housing). 281. There is language in James that suggested that the Court saw low-income individuals as different because they did not form a permanent, identifiable group but were rather one of many "diverse and shifting groups that make up the Ameri- can people." Id. (emphasis added); see also Niblock, supra note 180, at 187 (noting that many commentators treat James "as an aberration, and not as a guide to how the Court should evaluate laws placing discriminatory political burdens on identifiable groups") (citation omitted). It is also important to keep in mind that James preceded Davis v. Bandeiner, 478 U.S. 109 (1986), a decision in which a plurality of Justices indicated that the systematic exclusion of a political party from equal access to the political process would present a constitutional problem, Davis, 478 U.S. at 132. 1997] ROMER V. EVANS not confined to cases involving initiatives or referendum votes. It also lies at the heart of the reapportionment cases where the mandated equality is unrelated to the individual's membership in a suspect class. 282 The Court's position with regard to political ger- rymandering is also relevant. Though the state has authority to draw district lines, it may not do so in a way that systematically "degrade[s] ...a group of voters' influence on the political process as a whole."283

B. The PoliticalDimension-Advantages of the FundamentalRights Approach The fundamental rights approach has a number of advantages over the enhanced rational basis review which the Court used in Romer. Many voters in Colorado supported Amendment 2 because they believed that it prevented gays and lesbians from obtaining special advantages. 284 This characterization of Amend- ment 2 was dismissed by the Court as "implausible."285 As a re- 286 sult, the Court was accused of being elitist and undemocratic. The accusation with the greatest potential of doing political damage to the Court was the charge that it had usurped the people's rightful role. 287 The political danger is well illustrated in an editorial in the Chicago Tribune, which claims both sympathy for homosexuals and a lack thereof for the Court's action. 288 Even though the newspaper admitted that "homosexuals have been badly treated for most of American history and that the law should try to rectify that injustice," 289 it asserted that Romer was "a defeat

282. See Reynolds v. Simms, 377 U.S. 533 (1964) (holding unconstitutional Ala- bama's state legislative apportionment, and in so doing, introducing the one- person, one-vote standard and noting that vote dilution debased democracy). 283. Davis, 478 U.S. at 110. 284. Marcosson, supra note 25, at 138 n.6. 285. Evans v. Romer, 116 S. Ct. 1620, 1624 (1996). 286. See id. at 1629 (Scalia, J., dissenting). 287. See supra notes 40-43 and accompanying text. Rhetoric matters in politics; it affects how people think about problems. See MURRAY EDELMAN, THE SYMBOLIC USES OF POLITICS 114-30 (1985) (examining how language affects politics). "Democracy" is a word of some particular potency. See also ROBERT A. DAHL, DEMOCRACY AND ITS CRITICS 2 (1989). The notion of majority rule is particularly fundamental in the United States, though it at times runs counter to belief in individual rights. The charge that the people's power has been usurped thus threat- ens the legitimacy of the action and the agency that took it. 288. The Supreme Court Overreaches,supra note 17, at 16. 289. Id. Law and Inequality [Vol. 15:275 for democratic governance that may ultimately inflame rather 290 than erode opposition to gay civil rights." Democracy is a word of considerable political potency. It is generally accepted by Americans that the people should in some sense rule, and that actions against the majority carry a presumption of illegitimacy. However, it is not always easy to reconcile the various rights, including minority rights, which we consider to be part of democratic government. Nor is it easy to determine which majority in a particular situation should be controlling. In Romer, for example, the city ordinances of Aspen, Denver and Boulder protecting homosexuals from discrimination and repealed by Amendment 2 were passed by normal democratic means.291 It certainly could be argued that, at least for those localities, anti- discrimination laws represented the true will of the people. But the scope of a conflict often determines its outcome. The defeated have an incentive to bring in other participants to shift the balance of power. 292 In Colorado the loss in some arenas, brought about by local majorities, led anti-gay forces to seek a statewide majority with a different political composition. Con- ceivably a still larger majority might someday be found, composed of the nation at large, which could if it chose enact national civil rights laws overruling legislation at the state level. Political struc- tures establish a hierarchy of power, with larger majorities domi- nating smaller ones, but there is no moral imperative within democratic theory that this must be the case. While democratic theory asserts that "the people" must in some sense govern or hold those who do govern accountable, the theory does not define who "the people" should be. Even as decided, Romer was not really undemocratic. The will of local majorities was upheld in the Court. It could be argued as well, that to the extent that the campaign for Amendment 2 misled voters, the vote was not really a true reflection of what the people of Colorado wished. 293 However, there is a political dimen- sion to U.S. Supreme Court decisions, as discussed earlier, and the influence of the Court is weakened when it loses public confidence

290. Id. 291. See supra notes 40-43 and accompanying text (discussing the passage of Amendment 2 by referendum and the importance of this fact in the litigation). 292. E.E. SCHATTSNEIDER, THE SEMI-SOVEREIGN PEOPLE 1-19 (1960) (discussing the difficulty in taking such steps to balance the power dynamic if one lacks the necessary resources and support). 293. I do not mean to suggest by this that votes ever accurately reflect majority sentiment. 1997] ROMER V. EVANS in the legitimacy of its decisions. The use of the fundamental rights approach in Romer would have been clearer to the public than the reasoning actually employed by the Court. Moreover, it would have been just as valid as a matter of constitutional interpretation. Most importantly, the fundamental rights approach would have emphasized the Court's commitment to democratic processes and would have blunted the fact that it had overruled a statewide referendum. In addition, the fundamental rights approach would have avoided a number of other legal and political problems in the Court's equal protection analysis. In holding that the State could not act out of "animus," the Court invited future litigation based on the motivation behind laws, which is often very difficult to determine. 294 The fundamental rights approach would have avoided such questions. The fundamental rights approach does not involve homosexual sexual conduct at all, and so the Bowers opinion used by the dissent would no longer be relevant-295 Finally, such an approach would have allowed the Court to avoid the charge that it had taken sides in the "culture wars,"296 since it would merely have been defending the openness of the democratic process.

Conclusion The Supreme Court's decision in Romer v.Evans was correct in its outcome. The Court acted clearly within the tradition of keeping the political system fair. Amendment 2 deliberately singled out gays, lesbians and bisexuals regardless of their actual conduct, and placed them at a disadvantage in obtaining the benefits of society that other citizens take for granted. It was correctly overturned because it denied homosexuals the equal protection of the laws guaranteed by the Constitution. Romer could have been argued more convincingly, however, either by providing more explicit support for the use of the enhanced rational basis test or by using a standard of review based upon fundamental rights. Not only would the Court have been better able to defend its holding against legal attacks, such as those raised by Justice Scalia in the dissent, but it also would have

294. Relying on impermissible motivations has been widely criticized at various times by many commentators and Justices. The criticism made here is not confined to those who hold any particular interpretative perspective. See, e.g., TRIBE, supra note 50, § 12-6, at 821-25 (indicating some of the issues and objections that have been leveled). 295. See supra notes 165-71 and accompanying text; see also Part III.D. 296. Evans v. Romer, 116 S. Ct. 1620, 1629 (1996). Law and Iiequality [Vol. 15:275 been better immunized from public criticism. In this high-profile and politically important case, it was particularly important that the public not come away with the notion that an amendment which purported only to bar "special rights" and "preferences," and which had been voted on directly by the people of Colorado in a referendum, was then blocked by an elitist Court promoting its own policy views. At the least, Justice Kennedy should have confronted the "special rights" argument more directly, in terms more easily understood by the public and by the news media. At the same time, it is important to remember that any political damage to the Court caused by Romer is likely to be limited. The Court's action, while invalidating a citizen referendum, still showed support for a form of majority rule. It also did not step beyond how most Americans believe that gays, lesbians and bisexuals should be treated.2 97 Romer was also a narrow decision, despite the broad reach of the amendment which was struck down. Specifically, the Court was urged by gay-rights supporters to declare sexual orientation to be a suspect classification, and it did not do so. If it had taken this course, its decision would have been 29 "clearer-and more controversial," 8 and would have resulted in its injecting itself directly into such emotional and politically charged "social" issues as single-sex marriage. The Court may well have wished to avoid such controversy, at least for the present.

297. Data show that there is strong public support for homosexuals regarding equal access to jobs, housing and public accommodations, but opposition on other issues such as same-sex marriage. See Marcosson, supra note 25, at 138 n.6 (discussing the language used in Amendment 2 and the public's reaction to some of its specific components). 298. Greenburg, supra note 17, at 4.